2020
|
| Justo, Gabriel; Hoffmann, Renato Barreto; Vogel, Adriano; Griebler, Dalvan; Fernandes, Luiz G. L. Acelerando uma Aplicação de Detecção de Pistas com MPI Inproceedings doi In: XX Escola Regional de Alto Desempenho da Região Sul (ERAD-RS), pp. 117-120, Sociedade Brasileira de Computação (SBC), Santa Maria, BR, 2020. @inproceedings{JUSTO:ERAD:20,
title = {Acelerando uma Aplicação de Detecção de Pistas com MPI},
author = {Gabriel Justo and Renato Barreto Hoffmann and Adriano Vogel and Dalvan Griebler and Luiz G. L. Fernandes},
url = {https://doi.org/10.5753/eradrs.2020.10770},
doi = {10.5753/eradrs.2020.10770},
year = {2020},
date = {2020-04-01},
booktitle = {XX Escola Regional de Alto Desempenho da Região Sul (ERAD-RS)},
pages = {117-120},
publisher = {Sociedade Brasileira de Computação (SBC)},
address = {Santa Maria, BR},
abstract = {Aplicações de stream de vídeo demandam processamento de alto desempenho para atender requisitos de tempo real. Nesse cenário, a programação paralela distribuída é uma alternativa para acelerar e escalar o desempenho. Neste trabalho, o objetivo é paralelizar uma aplicação de detecção de pistas com a biblioteca MPI usando o padrão Farm e implementando duas estratégias de distribuição de tarefas. Os resultados evidenciam os ganhos de desempenho.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Aplicações de stream de vídeo demandam processamento de alto desempenho para atender requisitos de tempo real. Nesse cenário, a programação paralela distribuída é uma alternativa para acelerar e escalar o desempenho. Neste trabalho, o objetivo é paralelizar uma aplicação de detecção de pistas com a biblioteca MPI usando o padrão Farm e implementando duas estratégias de distribuição de tarefas. Os resultados evidenciam os ganhos de desempenho. |
| Araujo, Gabriell Alves; Griebler, Dalvan; Danelutto, Marco; Fernandes, Luiz Gustavo Efficient NAS Parallel Benchmark Kernels with CUDA Inproceedings doi In: 28th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP), pp. 9-16, IEEE, Västerås, Sweden, Sweden, 2020. @inproceedings{ARAUJO:PDP:20,
title = {Efficient NAS Parallel Benchmark Kernels with CUDA},
author = {Gabriell Alves Araujo and Dalvan Griebler and Marco Danelutto and Luiz Gustavo Fernandes},
url = {https://doi.org/10.1109/PDP50117.2020.00009},
doi = {10.1109/PDP50117.2020.00009},
year = {2020},
date = {2020-03-01},
booktitle = {28th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP)},
pages = {9-16},
publisher = {IEEE},
address = {Västerås, Sweden, Sweden},
series = {PDP'20},
abstract = {NAS Parallel Benchmarks (NPB) are one of the standard benchmark suites used to evaluate parallel hardware and software. There are many research efforts trying to provide different parallel versions apart from the original OpenMP and MPI. Concerning GPU accelerators, there are only the OpenCL and OpenACC available as consolidated versions. Our goal is to provide an efficient parallel implementation of the five NPB kernels with CUDA. Our contribution covers different aspects. First, best parallel programming practices were followed to implement NPB kernels using CUDA. Second, the support of larger workloads (class B and C) allow to stress and investigate the memory of robust GPUs. Third, we show that it is possible to make NPB efficient and suitable for GPUs although the benchmarks were designed for CPUs in the past. We succeed in achieving double performance with respect to the state-of-the-art in some cases as well as implementing efficient memory usage. Fourth, we discuss new experiments comparing performance and memory usage against OpenACC and OpenCL state-of-the-art versions using a relative new GPU architecture. The experimental results also revealed that our version is the best one for all the NPB kernels compared to OpenACC and OpenCL. The greatest differences were observed for the FT and EP kernels.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
NAS Parallel Benchmarks (NPB) are one of the standard benchmark suites used to evaluate parallel hardware and software. There are many research efforts trying to provide different parallel versions apart from the original OpenMP and MPI. Concerning GPU accelerators, there are only the OpenCL and OpenACC available as consolidated versions. Our goal is to provide an efficient parallel implementation of the five NPB kernels with CUDA. Our contribution covers different aspects. First, best parallel programming practices were followed to implement NPB kernels using CUDA. Second, the support of larger workloads (class B and C) allow to stress and investigate the memory of robust GPUs. Third, we show that it is possible to make NPB efficient and suitable for GPUs although the benchmarks were designed for CPUs in the past. We succeed in achieving double performance with respect to the state-of-the-art in some cases as well as implementing efficient memory usage. Fourth, we discuss new experiments comparing performance and memory usage against OpenACC and OpenCL state-of-the-art versions using a relative new GPU architecture. The experimental results also revealed that our version is the best one for all the NPB kernels compared to OpenACC and OpenCL. The greatest differences were observed for the FT and EP kernels. |
| Vogel, Adriano; Rista, Cassiano; Justo, Gabriel; Ewald, Endrius; Griebler, Dalvan; Mencagli, Gabriele; Fernandes, Luiz Gustavo Parallel Stream Processing with MPI for Video Analytics and Data Visualization Inproceedings doi In: High Performance Computing Systems, pp. 102-116, Springer, Cham, 2020. @inproceedings{VOGEL:CCIS:20,
title = {Parallel Stream Processing with MPI for Video Analytics and Data Visualization},
author = {Adriano Vogel and Cassiano Rista and Gabriel Justo and Endrius Ewald and Dalvan Griebler and Gabriele Mencagli and Luiz Gustavo Fernandes},
url = {https://doi.org/10.1007/978-3-030-41050-6_7},
doi = {10.1007/978-3-030-41050-6_7},
year = {2020},
date = {2020-02-01},
booktitle = {High Performance Computing Systems},
volume = {1171},
pages = {102-116},
publisher = {Springer},
address = {Cham},
series = {Communications in Computer and Information Science (CCIS)},
abstract = {The amount of data generated is increasing exponentially. However, processing data and producing fast results is a technological challenge. Parallel stream processing can be implemented for handling high frequency and big data flows. The MPI parallel programming model offers low-level and flexible mechanisms for dealing with distributed architectures such as clusters. This paper aims to use it to accelerate video analytics and data visualization applications so that insight can be obtained as soon as the data arrives. Experiments were conducted with a Domain-Specific Language for Geospatial Data Visualization and a Person Recognizer video application. We applied the same stream parallelism strategy and two task distribution strategies. The dynamic task distribution achieved better performance than the static distribution in the HPC cluster. The data visualization achieved lower throughput with respect to the video analytics due to the I/O intensive operations. Also, the MPI programming model shows promising performance outcomes for stream processing applications.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The amount of data generated is increasing exponentially. However, processing data and producing fast results is a technological challenge. Parallel stream processing can be implemented for handling high frequency and big data flows. The MPI parallel programming model offers low-level and flexible mechanisms for dealing with distributed architectures such as clusters. This paper aims to use it to accelerate video analytics and data visualization applications so that insight can be obtained as soon as the data arrives. Experiments were conducted with a Domain-Specific Language for Geospatial Data Visualization and a Person Recognizer video application. We applied the same stream parallelism strategy and two task distribution strategies. The dynamic task distribution achieved better performance than the static distribution in the HPC cluster. The data visualization achieved lower throughput with respect to the video analytics due to the I/O intensive operations. Also, the MPI programming model shows promising performance outcomes for stream processing applications. |
2019
|
| Pieper, Ricardo; Griebler, Dalvan; Fernandes, Luiz G. Structured Stream Parallelism for Rust Inproceedings doi In: XXIII Brazilian Symposium on Programming Languages (SBLP), pp. 54-61, ACM, Salvador, Brazil, 2019. @inproceedings{PIEPER:SBLP:19,
title = {Structured Stream Parallelism for Rust},
author = {Ricardo Pieper and Dalvan Griebler and Luiz G. Fernandes},
url = {https://doi.org/10.1145/3355378.3355384},
doi = {10.1145/3355378.3355384},
year = {2019},
date = {2019-10-01},
booktitle = {XXIII Brazilian Symposium on Programming Languages (SBLP)},
pages = {54-61},
publisher = {ACM},
address = {Salvador, Brazil},
series = {SBLP'19},
abstract = {Structured parallel programming has been studied and applied in several programming languages. This approach has proven to be suitable for abstracting low-level and architecture-dependent parallelism implementations. Our goal is to provide a structured and high-level library for the Rust language, targeting parallel stream processing applications for multi-core servers. Rust is an emerging programming language that has been developed by Mozilla Research group, focusing on performance, memory safety, and thread-safety. However, it lacks parallel programming abstractions, especially for stream processing applications. This paper contributes to a new API based on the structured parallel programming approach to simplify parallel software developing. Our experiments highlight that our solution provides higher-level parallel programming abstractions for stream processing applications in Rust. We also show that the throughput and speedup are comparable to the state-of-the-art for certain workloads.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Structured parallel programming has been studied and applied in several programming languages. This approach has proven to be suitable for abstracting low-level and architecture-dependent parallelism implementations. Our goal is to provide a structured and high-level library for the Rust language, targeting parallel stream processing applications for multi-core servers. Rust is an emerging programming language that has been developed by Mozilla Research group, focusing on performance, memory safety, and thread-safety. However, it lacks parallel programming abstractions, especially for stream processing applications. This paper contributes to a new API based on the structured parallel programming approach to simplify parallel software developing. Our experiments highlight that our solution provides higher-level parallel programming abstractions for stream processing applications in Rust. We also show that the throughput and speedup are comparable to the state-of-the-art for certain workloads. |
| Maliszewski, Anderson; Roloff, Eduardo; Griebler, Dalvan; Navaux, Philippe O Impacto da Interconexão de Rede no Desempenho de Programas Paralelos Inproceedings doi In: Anais do XX Simpósio em Sistemas Computacionais de Alto Desempenho, pp. 73-84, Sociedade Brasileira de Computação, Campo Grande, Brazil, 2019. @inproceedings{larcc:impacto_interconexao_HPC:WSCAD:19,
title = {O Impacto da Interconexão de Rede no Desempenho de Programas Paralelos},
author = {Anderson Maliszewski and Eduardo Roloff and Dalvan Griebler and Philippe Navaux},
url = {https://doi.org/10.5753/wscad.2019.8658},
doi = {10.5753/wscad.2019.8658},
year = {2019},
date = {2019-10-01},
booktitle = {Anais do XX Simpósio em Sistemas Computacionais de Alto Desempenho},
pages = {73-84},
publisher = {Sociedade Brasileira de Computação},
address = {Campo Grande, Brazil},
series = {WSCAD'19},
abstract = {O desempenho de aplicações paralelas depende de dois componentes principais do ambiente; o poder de processamento e a interconexão de rede. Neste trabalho, foi avaliado o impacto de uma interconexão de alto desempenho em programas paralelos em um cluster homogêneo de servidores interconectados por Gigabit Ethernet 1 Gbps e InfiniBand FDR 56 Gbps. Foi realizada uma caracterização do NAS Parallel Benchmarks em relação à computação, comunicação e custo de execução em instâncias da Microsoft Azure. Os resultados mostraram que, em aplicações altamente dependentes de rede, o desempenho pode ser significativamente melhorado ao utilizar InfiniBand a um custo de execução melhor, mesmo com o preço superior da instância.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
O desempenho de aplicações paralelas depende de dois componentes principais do ambiente; o poder de processamento e a interconexão de rede. Neste trabalho, foi avaliado o impacto de uma interconexão de alto desempenho em programas paralelos em um cluster homogêneo de servidores interconectados por Gigabit Ethernet 1 Gbps e InfiniBand FDR 56 Gbps. Foi realizada uma caracterização do NAS Parallel Benchmarks em relação à computação, comunicação e custo de execução em instâncias da Microsoft Azure. Os resultados mostraram que, em aplicações altamente dependentes de rede, o desempenho pode ser significativamente melhorado ao utilizar InfiniBand a um custo de execução melhor, mesmo com o preço superior da instância. |
 | Mencagli, Gabriele; Torquati, Massimo; Griebler, Dalvan; Danelutto, Marco; Fernandes, Luiz Gustavo L. Raising the Parallel Abstraction Level for Streaming Analytics Applications Journal Article doi In: IEEE Access, vol. 7, pp. 131944 - 131961, 2019. @article{MENCAGLI:IEEEAccess:19,
title = {Raising the Parallel Abstraction Level for Streaming Analytics Applications},
author = {Gabriele Mencagli and Massimo Torquati and Dalvan Griebler and Marco Danelutto and Luiz Gustavo L. Fernandes},
url = {https://doi.org/10.1109/ACCESS.2019.2941183},
doi = {10.1109/ACCESS.2019.2941183},
year = {2019},
date = {2019-09-01},
journal = {IEEE Access},
volume = {7},
pages = {131944 - 131961},
publisher = {IEEE},
abstract = {In the stream processing domain, applications are represented by graphs of operators arbitrarily connected and filled with their business logic code. The APIs of existing Stream Processing Systems (SPSs) ease the development of transformations that recur in the streaming practice (e.g., filtering, aggregation and joins). In contrast, their parallelism abstractions are quite limited since they provide support to stateless operators only, or when the state is organized in a set of key-value pairs. This paper presents how the parallel patterns methodology can be revisited for sliding-window streaming analytics. Our vision fosters a design process of the application as composition and nesting of ready-to-use patterns provided through a C++17 fluent interface. Our prototype implements the run-time system of the patterns in the FastFlow parallel library expressing thread-based parallelism. The experimental analysis shows interesting outcomes. First, our pattern-based approach allows easy prototyping of different versions of the application, and the programmer can leverage nesting of patterns to increase performance (up to 37% in one of the two considered test-bed cases). Second, our FastFlow implementation outperforms (three times faster) the handmade porting of our patterns in popular JVM-based SPSs. Finally, in the concluding part of this paper, we explore the use of a task-based run-time system, by deriving interesting insights into how to make our patterns library suitable for multi backends.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In the stream processing domain, applications are represented by graphs of operators arbitrarily connected and filled with their business logic code. The APIs of existing Stream Processing Systems (SPSs) ease the development of transformations that recur in the streaming practice (e.g., filtering, aggregation and joins). In contrast, their parallelism abstractions are quite limited since they provide support to stateless operators only, or when the state is organized in a set of key-value pairs. This paper presents how the parallel patterns methodology can be revisited for sliding-window streaming analytics. Our vision fosters a design process of the application as composition and nesting of ready-to-use patterns provided through a C++17 fluent interface. Our prototype implements the run-time system of the patterns in the FastFlow parallel library expressing thread-based parallelism. The experimental analysis shows interesting outcomes. First, our pattern-based approach allows easy prototyping of different versions of the application, and the programmer can leverage nesting of patterns to increase performance (up to 37% in one of the two considered test-bed cases). Second, our FastFlow implementation outperforms (three times faster) the handmade porting of our patterns in popular JVM-based SPSs. Finally, in the concluding part of this paper, we explore the use of a task-based run-time system, by deriving interesting insights into how to make our patterns library suitable for multi backends. |
 | Fischer, Gabriel Souto; Righi, Rodrigo Rosa; Costa, Cristiano André; Galante, Guilherme; Griebler, Dalvan Towards Evaluating Proactive and Reactive Approaches on Reorganizing Human Resources in IoT-Based Smart Hospitals Journal Article doi In: Sensors, vol. 19, no. 17, pp. 3800, 2019. @article{FISHER:Elasticity-Hospital:SENSORS:19,
title = {Towards Evaluating Proactive and Reactive Approaches on Reorganizing Human Resources in IoT-Based Smart Hospitals},
author = {Gabriel Souto Fischer and Rodrigo Rosa Righi and Cristiano André Costa and Guilherme Galante and Dalvan Griebler},
url = {https://doi.org/10.3390/s19173800},
doi = {10.3390/s19173800},
year = {2019},
date = {2019-09-01},
urldate = {2019-09-01},
journal = {Sensors},
volume = {19},
number = {17},
pages = {3800},
publisher = {MDPI},
abstract = {Hospitals play an important role on ensuring a proper treatment of human health. One of the problems to be faced is the increasingly overcrowded patients care queues, who end up waiting for longer times without proper treatment to their health problems. The allocation of health professionals in hospital environments is not able to adapt to the demands of patients. There are times when underused rooms have idle professionals, and overused rooms have fewer professionals than necessary. Previous works have not solved this problem since they focus on understanding the evolution of doctor supply and patient demand, as to better adjust one to the other. However, they have not proposed concrete solutions for that regarding techniques for better allocating available human resources. Moreover, elasticity is one of the most important features of cloud computing, referring to the ability to add or remove resources according to the needs of the application or service. Based on this background, we introduce Elastic allocation of human resources in Healthcare environments (ElHealth) an IoT-focused model able to monitor patient usage of hospital rooms and adapt these rooms for patients demand. Using reactive and proactive elasticity approaches, ElHealth identifies when a room will have a demand that exceeds the capacity of care, and proposes actions to move human resources to adapt to patient demand. Our main contribution is the definition of Human Resources IoT-based Elasticity (i.e., an extension of the concept of resource elasticity in Cloud Computing to manage the use of human resources in a healthcare environment, where health professionals are allocated and deallocated according to patient demand). Another contribution is a cost–benefit analysis for the use of reactive and predictive strategies on human resources reorganization. ElHealth was simulated on a hospital environment using data from a Brazilian polyclinic, and obtained promising results, decreasing the waiting time by up to 96.4% and 96.73% in reactive and proactive approaches, respectively.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Hospitals play an important role on ensuring a proper treatment of human health. One of the problems to be faced is the increasingly overcrowded patients care queues, who end up waiting for longer times without proper treatment to their health problems. The allocation of health professionals in hospital environments is not able to adapt to the demands of patients. There are times when underused rooms have idle professionals, and overused rooms have fewer professionals than necessary. Previous works have not solved this problem since they focus on understanding the evolution of doctor supply and patient demand, as to better adjust one to the other. However, they have not proposed concrete solutions for that regarding techniques for better allocating available human resources. Moreover, elasticity is one of the most important features of cloud computing, referring to the ability to add or remove resources according to the needs of the application or service. Based on this background, we introduce Elastic allocation of human resources in Healthcare environments (ElHealth) an IoT-focused model able to monitor patient usage of hospital rooms and adapt these rooms for patients demand. Using reactive and proactive elasticity approaches, ElHealth identifies when a room will have a demand that exceeds the capacity of care, and proposes actions to move human resources to adapt to patient demand. Our main contribution is the definition of Human Resources IoT-based Elasticity (i.e., an extension of the concept of resource elasticity in Cloud Computing to manage the use of human resources in a healthcare environment, where health professionals are allocated and deallocated according to patient demand). Another contribution is a cost–benefit analysis for the use of reactive and predictive strategies on human resources reorganization. ElHealth was simulated on a hospital environment using data from a Brazilian polyclinic, and obtained promising results, decreasing the waiting time by up to 96.4% and 96.73% in reactive and proactive approaches, respectively. |
| Rockenbach, Dinei A.; Griebler, Dalvan; Danelutto, Marco; Fernandes, Luiz Gustavo High-Level Stream Parallelism Abstractions with SPar Targeting GPUs Inproceedings doi In: Parallel Computing is Everywhere, Proceedings of the International Conference on Parallel Computing (ParCo), pp. 543-552, IOS Press, Prague, Czech Republic, 2019. @inproceedings{ROCKENBACH:PARCO:19,
title = {High-Level Stream Parallelism Abstractions with SPar Targeting GPUs},
author = {Dinei A. Rockenbach and Dalvan Griebler and Marco Danelutto and Luiz Gustavo Fernandes},
url = {https://doi.org/10.3233/APC200083},
doi = {10.3233/APC200083},
year = {2019},
date = {2019-09-01},
booktitle = {Parallel Computing is Everywhere, Proceedings of the International Conference on Parallel Computing (ParCo)},
volume = {36},
pages = {543-552},
publisher = {IOS Press},
address = {Prague, Czech Republic},
series = {ParCo'19},
abstract = {The combined exploitation of stream and data parallelism is demonstrating encouraging performance results in the literature for heterogeneous architectures, which are present on every computer systems today. However, provide parallel software efficiently targeting those architectures requires significant programming effort and expertise. The SPar domain-specific language already represents a solution to this problem providing proven high-level programming abstractions for multi-core architectures. In this paper, we enrich the SPar language adding support for GPUs. New transformation rules are designed for generating parallel code using stream and data parallel patterns. Our experiments revealed that these transformations rules are able to improve performance while the high-level programming abstractions are maintained.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The combined exploitation of stream and data parallelism is demonstrating encouraging performance results in the literature for heterogeneous architectures, which are present on every computer systems today. However, provide parallel software efficiently targeting those architectures requires significant programming effort and expertise. The SPar domain-specific language already represents a solution to this problem providing proven high-level programming abstractions for multi-core architectures. In this paper, we enrich the SPar language adding support for GPUs. New transformation rules are designed for generating parallel code using stream and data parallel patterns. Our experiments revealed that these transformations rules are able to improve performance while the high-level programming abstractions are maintained. |
| Vogel, Adriano; Griebler, Dalvan; Danelutto, Marco; Fernandes, Luiz Gustavo Seamless Parallelism Management for Multi-core Stream Processing Inproceedings doi In: Advances in Parallel Computing, Proceedings of the International Conference on Parallel Computing (ParCo), pp. 533-542, IOS Press, Prague, Czech Republic, 2019. @inproceedings{VOGEL:PARCO:19,
title = {Seamless Parallelism Management for Multi-core Stream Processing},
author = {Adriano Vogel and Dalvan Griebler and Marco Danelutto and Luiz Gustavo Fernandes},
url = {https://doi.org/10.3233/APC200082},
doi = {10.3233/APC200082},
year = {2019},
date = {2019-09-01},
booktitle = {Advances in Parallel Computing, Proceedings of the International Conference on Parallel Computing (ParCo)},
volume = {36},
pages = {533-542},
publisher = {IOS Press},
address = {Prague, Czech Republic},
series = {ParCo'19},
abstract = {Video streaming applications have critical performance requirements for dealing with fluctuating workloads and providing results in real-time. As a consequence, the majority of these applications demand parallelism for delivering quality of service to users. Although high-level and structured parallel programming aims at facilitating parallelism exploitation, there are still several issues to be addressed for increasing/improving existing parallel programming abstractions. In this paper, we aim at employing self-adaptivity for stream processing in order to seamlessly manage the application parallelism configurations at run-time, where a new strategy alleviates from application programmers the need to set time-consuming and error-prone parallelism parameters. The new strategy was implemented and validated on SPar. The results have shown that the proposed solution increases the level of abstraction and achieved a competitive performance.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Video streaming applications have critical performance requirements for dealing with fluctuating workloads and providing results in real-time. As a consequence, the majority of these applications demand parallelism for delivering quality of service to users. Although high-level and structured parallel programming aims at facilitating parallelism exploitation, there are still several issues to be addressed for increasing/improving existing parallel programming abstractions. In this paper, we aim at employing self-adaptivity for stream processing in order to seamlessly manage the application parallelism configurations at run-time, where a new strategy alleviates from application programmers the need to set time-consuming and error-prone parallelism parameters. The new strategy was implemented and validated on SPar. The results have shown that the proposed solution increases the level of abstraction and achieved a competitive performance. |
| Teixeira, Djalma; Vogel, Adriano; Griebler, Dalvan Proposta de Monitoramento e Gerenciamento Inteligente de Temperatura em Datacenters Inproceedings In: 16th Escola Regional de Redes de Computadores (ERRC), pp. 1-8, Sociedade Brasileira de Computação, Alegrete, Brazil, 2019. @inproceedings{larcc:smart_datacenter_temperatura:ERRC:19,
title = {Proposta de Monitoramento e Gerenciamento Inteligente de Temperatura em Datacenters},
author = {Djalma Teixeira and Adriano Vogel and Dalvan Griebler},
url = {https://sol.sbc.org.br/index.php/errc/article/view/9209/9112},
year = {2019},
date = {2019-09-01},
booktitle = {16th Escola Regional de Redes de Computadores (ERRC)},
pages = {1-8},
publisher = {Sociedade Brasileira de Computação},
address = {Alegrete, Brazil},
series = {ERRC'19},
abstract = {O aumento constante do crescimento e desenvolvimento das infraestruturas computacionais, vem impulsionando uma demanda cada vez maior por monitoramento e gerenciamento inteligente de datacenters. Em um ambiente gerenciado autonomicamente os equipamentos são controlados por meio de ações autonômicas, que são executadas sob determinadas condições sem a necessidade de intervenção humana. O objetivo deste trabalho é propor um modelo conceitual de monitoramento e gerenciamento inteligente para temperatura, que pode ser aplicado tanto em estruturas básicas quanto complexas e adaptado a heterogeneidade dos datacenters atuais.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
O aumento constante do crescimento e desenvolvimento das infraestruturas computacionais, vem impulsionando uma demanda cada vez maior por monitoramento e gerenciamento inteligente de datacenters. Em um ambiente gerenciado autonomicamente os equipamentos são controlados por meio de ações autonômicas, que são executadas sob determinadas condições sem a necessidade de intervenção humana. O objetivo deste trabalho é propor um modelo conceitual de monitoramento e gerenciamento inteligente para temperatura, que pode ser aplicado tanto em estruturas básicas quanto complexas e adaptado a heterogeneidade dos datacenters atuais. |
| Vogel, Adriano; Griebler, Dalvan; Danelutto, Marco; Fernandes, Luiz Gustavo Minimizing Self-Adaptation Overhead in Parallel Stream Processing for Multi-Cores Inproceedings doi In: Euro-Par 2019: Parallel Processing Workshops, pp. 12, Springer, Göttingen, Germany, 2019. @inproceedings{VOGEL:adaptive-overhead:AutoDaSP:19,
title = {Minimizing Self-Adaptation Overhead in Parallel Stream Processing for Multi-Cores},
author = {Adriano Vogel and Dalvan Griebler and Marco Danelutto and Luiz Gustavo Fernandes},
url = {https://doi.org/10.1007/978-3-030-48340-1_3},
doi = {10.1007/978-3-030-48340-1_3},
year = {2019},
date = {2019-08-01},
booktitle = {Euro-Par 2019: Parallel Processing Workshops},
volume = {11997},
pages = {12},
publisher = {Springer},
address = {Göttingen, Germany},
series = {Lecture Notes in Computer Science},
abstract = {Stream processing paradigm is present in several applications that apply computations over continuous data flowing in the form of streams (e.g., video feeds, image, and data analytics). Employing self-adaptivity to stream processing applications can provide higher-level programming abstractions and autonomic resource management. However, there are cases where the performance is suboptimal. In this paper, the goal is to optimize parallelism adaptations in terms of stability and accuracy, which can improve the performance of parallel stream processing applications. Therefore, we present a new optimized self-adaptive strategy that is experimentally evaluated. The proposed solution provided high-level programming abstractions, reduced the adaptation overhead, and achieved a competitive performance with the best static executions.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Stream processing paradigm is present in several applications that apply computations over continuous data flowing in the form of streams (e.g., video feeds, image, and data analytics). Employing self-adaptivity to stream processing applications can provide higher-level programming abstractions and autonomic resource management. However, there are cases where the performance is suboptimal. In this paper, the goal is to optimize parallelism adaptations in terms of stability and accuracy, which can improve the performance of parallel stream processing applications. Therefore, we present a new optimized self-adaptive strategy that is experimentally evaluated. The proposed solution provided high-level programming abstractions, reduced the adaptation overhead, and achieved a competitive performance with the best static executions. |
| Maliszewski, Anderson M.; Vogel, Adriano; Griebler, Dalvan; Roloff, Eduardo; Fernandes, Luz G.; Navaux, Philippe O. A. Minimizing Communication Overheads in Container-based Clouds for HPC Applications Inproceedings doi In: IEEE Symposium on Computers and Communications (ISCC), pp. 1-6, IEEE, Barcelona, Spain, 2019. @inproceedings{larcc:communication_overhead_lxd:ISCC:19,
title = {Minimizing Communication Overheads in Container-based Clouds for HPC Applications},
author = {Anderson M. Maliszewski and Adriano Vogel and Dalvan Griebler and Eduardo Roloff and Luz G. Fernandes and Philippe O. A. Navaux},
url = {https://doi.org/10.1109/ISCC47284.2019.8969716},
doi = {10.1109/ISCC47284.2019.8969716},
year = {2019},
date = {2019-07-01},
booktitle = {IEEE Symposium on Computers and Communications (ISCC)},
pages = {1-6},
publisher = {IEEE},
address = {Barcelona, Spain},
series = {ISCC'19},
abstract = {Although the industry has embraced the cloud computing model, there are still significant challenges to be addressed concerning the quality of cloud services. Network-intensive applications may not scale in the cloud due to the sharing of the network infrastructure. In the literature, performance evaluation studies are showing that the network tends to limit the scalability and performance of HPC applications. Therefore, we proposed the aggregation of Network Interface Cards (NICs) in a ready-to-use integration with the OpenNebula cloud manager using Linux containers. We perform a set of experiments using a network microbenchmark to get specific network performance metrics and NAS parallel benchmarks to analyze the performance impact on HPC applications. Our results highlight that the implementation of NIC aggregation improves network performance in terms of throughput and latency. Moreover, HPC applications have different patterns of behavior when using our approach, which depends on communication and the amount of data transferring. While network-intensive applications increased the performance up to 38%, other applications with aggregated NICs maintained the same performance or presented slightly worse performance.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Although the industry has embraced the cloud computing model, there are still significant challenges to be addressed concerning the quality of cloud services. Network-intensive applications may not scale in the cloud due to the sharing of the network infrastructure. In the literature, performance evaluation studies are showing that the network tends to limit the scalability and performance of HPC applications. Therefore, we proposed the aggregation of Network Interface Cards (NICs) in a ready-to-use integration with the OpenNebula cloud manager using Linux containers. We perform a set of experiments using a network microbenchmark to get specific network performance metrics and NAS parallel benchmarks to analyze the performance impact on HPC applications. Our results highlight that the implementation of NIC aggregation improves network performance in terms of throughput and latency. Moreover, HPC applications have different patterns of behavior when using our approach, which depends on communication and the amount of data transferring. While network-intensive applications increased the performance up to 38%, other applications with aggregated NICs maintained the same performance or presented slightly worse performance. |
 | Griebler, Dalvan; Vogel, Adriano; Sensi, Daniele De; Danelutto, Marco; Fernandes, Luiz Gustavo Simplifying and implementing service level objectives for stream parallelism Journal Article doi In: The Journal of Supercomputing, vol. 76, pp. 4603-4628, 2019, ISSN: 0920-8542. @article{GRIEBLER:JS:19,
title = {Simplifying and implementing service level objectives for stream parallelism},
author = {Dalvan Griebler and Adriano Vogel and Daniele De Sensi and Marco Danelutto and Luiz Gustavo Fernandes},
url = {https://doi.org/10.1007/s11227-019-02914-6},
doi = {10.1007/s11227-019-02914-6},
issn = {0920-8542},
year = {2019},
date = {2019-06-01},
urldate = {2019-06-01},
journal = {The Journal of Supercomputing},
volume = {76},
pages = {4603-4628},
publisher = {Springer},
abstract = {An increasing attention has been given to provide service level objectives (SLOs) in stream processing applications due to the performance and energy requirements, and because of the need to impose limits in terms of resource usage while improving the system utilization. Since the current and next-generation computing systems are intrinsically offering parallel architectures, the software has to naturally exploit the architecture parallelism. Implement and meet SLOs on existing applications is not a trivial task for application programmers, since the software development process, besides the parallelism exploitation, requires the implementation of autonomic algorithms or strategies. This is a system-oriented programming approach and requires the management of multiple knobs and sensors (e.g., the number of threads to use, the clock frequency of the cores, etc.) so that the system can self-adapt at runtime. In this work, we introduce a new and simpler way to define SLO in the application’s source code, by abstracting from the programmer all the details relative to self-adaptive system implementation. The application programmer specifies which parts of the code to parallelize and the related SLOs that should be enforced. To reach this goal, source-to-source code transformation rules are implemented in our compiler, which automatically generates self-adaptive strategies to enforce, at runtime, the user-expressed objectives. The experiments highlighted promising results with simpler, effective, and efficient SLO implementations for real-world applications.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
An increasing attention has been given to provide service level objectives (SLOs) in stream processing applications due to the performance and energy requirements, and because of the need to impose limits in terms of resource usage while improving the system utilization. Since the current and next-generation computing systems are intrinsically offering parallel architectures, the software has to naturally exploit the architecture parallelism. Implement and meet SLOs on existing applications is not a trivial task for application programmers, since the software development process, besides the parallelism exploitation, requires the implementation of autonomic algorithms or strategies. This is a system-oriented programming approach and requires the management of multiple knobs and sensors (e.g., the number of threads to use, the clock frequency of the cores, etc.) so that the system can self-adapt at runtime. In this work, we introduce a new and simpler way to define SLO in the application’s source code, by abstracting from the programmer all the details relative to self-adaptive system implementation. The application programmer specifies which parts of the code to parallelize and the related SLOs that should be enforced. To reach this goal, source-to-source code transformation rules are implemented in our compiler, which automatically generates self-adaptive strategies to enforce, at runtime, the user-expressed objectives. The experiments highlighted promising results with simpler, effective, and efficient SLO implementations for real-world applications. |
| Rockenbach, Dinei A.; Stein, Charles Michael; Griebler, Dalvan; Mencagli, Gabriele; Torquati, Massimo; Danelutto, Marco; Fernandes, Luiz Gustavo Stream Processing on Multi-cores with GPUs: Parallel Programming Models' Challenges Inproceedings doi In: International Parallel and Distributed Processing Symposium Workshops (IPDPSW), pp. 834-841, IEEE, Rio de Janeiro, Brazil, 2019. @inproceedings{ROCKENBACH:stream-multigpus:IPDPSW:19,
title = {Stream Processing on Multi-cores with GPUs: Parallel Programming Models' Challenges},
author = {Dinei A. Rockenbach and Charles Michael Stein and Dalvan Griebler and Gabriele Mencagli and Massimo Torquati and Marco Danelutto and Luiz Gustavo Fernandes},
url = {https://doi.org/10.1109/IPDPSW.2019.00137},
doi = {10.1109/IPDPSW.2019.00137},
year = {2019},
date = {2019-05-01},
booktitle = {International Parallel and Distributed Processing Symposium Workshops (IPDPSW)},
pages = {834-841},
publisher = {IEEE},
address = {Rio de Janeiro, Brazil},
series = {IPDPSW'19},
abstract = {The stream processing paradigm is used in several scientific and enterprise applications in order to continuously compute results out of data items coming from data sources such as sensors. The full exploitation of the potential parallelism offered by current heterogeneous multi-cores equipped with one or more GPUs is still a challenge in the context of stream processing applications. In this work, our main goal is to present the parallel programming challenges that the programmer has to face when exploiting CPUs and GPUs' parallelism at the same time using traditional programming models. We highlight the parallelization methodology in two use-cases (the Mandelbrot Streaming benchmark and the PARSEC's Dedup application) to demonstrate the issues and benefits of using heterogeneous parallel hardware. The experiments conducted demonstrate how a high-level parallel programming model targeting stream processing like the one offered by SPar can be used to reduce the programming effort still offering a good level of performance if compared with state-of-the-art programming models.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The stream processing paradigm is used in several scientific and enterprise applications in order to continuously compute results out of data items coming from data sources such as sensors. The full exploitation of the potential parallelism offered by current heterogeneous multi-cores equipped with one or more GPUs is still a challenge in the context of stream processing applications. In this work, our main goal is to present the parallel programming challenges that the programmer has to face when exploiting CPUs and GPUs' parallelism at the same time using traditional programming models. We highlight the parallelization methodology in two use-cases (the Mandelbrot Streaming benchmark and the PARSEC's Dedup application) to demonstrate the issues and benefits of using heterogeneous parallel hardware. The experiments conducted demonstrate how a high-level parallel programming model targeting stream processing like the one offered by SPar can be used to reduce the programming effort still offering a good level of performance if compared with state-of-the-art programming models. |
| Rockenbach, Dinei A.; Griebler, Dalvan; Fernandes, Luiz G. Proposta de Suporte ao Paralelismo de GPU na SPar Inproceedings In: Escola Regional de Alto Desempenho (ERAD-RS), pp. 4, Sociedade Brasileira de Computação (SBC), Três de Maio, BR, 2019. @inproceedings{ROCKENBACH:ERAD:19,
title = {Proposta de Suporte ao Paralelismo de GPU na SPar},
author = {Dinei A. Rockenbach and Dalvan Griebler and Luiz G. Fernandes},
url = {https://gmap.pucrs.br/dalvan/papers/2019/CR_ERAD_PG_Dinei_2019.pdf},
year = {2019},
date = {2019-04-01},
booktitle = {Escola Regional de Alto Desempenho (ERAD-RS)},
pages = {4},
publisher = {Sociedade Brasileira de Computação (SBC)},
address = {Três de Maio, BR},
abstract = {As GPUs (Graphics Processing Units) têm se destacado devido a seualto poder de processamento paralelo e sua presença crescente nos dispositivoscomputacionais. Porém, a sua exploração ainda requer conhecimento e esforçoconsideráveis do desenvolvedor. O presente trabalho propõe o suporte ao para-lelismo de GPU na SPar, que fornece um alto nível de abstração através de umalinguagem baseada em anotações do C++.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
As GPUs (Graphics Processing Units) têm se destacado devido a seualto poder de processamento paralelo e sua presença crescente nos dispositivoscomputacionais. Porém, a sua exploração ainda requer conhecimento e esforçoconsideráveis do desenvolvedor. O presente trabalho propõe o suporte ao para-lelismo de GPU na SPar, que fornece um alto nível de abstração através de umalinguagem baseada em anotações do C++. |
| Vogel, Adriano; Griebler, Dalvan; Fernandes, Luiz G. Adaptando o Paralelismo em Aplicações de Stream Conforme Objetivos de Throughput Inproceedings In: Escola Regional de Alto Desempenho (ERAD-RS), pp. 4, Sociedade Brasileira de Computação (SBC), Três de Maio, BR, 2019. @inproceedings{VOGEL:ERAD:19,
title = {Adaptando o Paralelismo em Aplicações de Stream Conforme Objetivos de Throughput},
author = {Adriano Vogel and Dalvan Griebler and Luiz G. Fernandes},
url = {https://gmap.pucrs.br/dalvan/papers/2019/CR_ERAD_PG_Vogel_2019.pdf},
year = {2019},
date = {2019-04-01},
booktitle = {Escola Regional de Alto Desempenho (ERAD-RS)},
pages = {4},
publisher = {Sociedade Brasileira de Computação (SBC)},
address = {Três de Maio, BR},
abstract = {As aplicações de processamento de streams possuem característicasde execuções dinâmicas com variações na carga e na demanda por recursos. Adaptar o grau de paralelismo é uma alternativa para responder a variaçãodurante a execução. Nesse trabalho é apresentada uma abstração de parale-lismo para a DSL SPar através de uma estratégia que autonomicamente adaptao grau de paralelismo de acordo com objetivos de desempenho.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
As aplicações de processamento de streams possuem característicasde execuções dinâmicas com variações na carga e na demanda por recursos. Adaptar o grau de paralelismo é uma alternativa para responder a variaçãodurante a execução. Nesse trabalho é apresentada uma abstração de parale-lismo para a DSL SPar através de uma estratégia que autonomicamente adaptao grau de paralelismo de acordo com objetivos de desempenho. |
| Maron, Carlos A. F.; Griebler, Dalvan; Fernandes, Luiz G. Benchmark Paramétrico para o Domínio do Paralelismo de Stream: Um Estudo de Caso com o Ferret da Suíte PARSEC Inproceedings In: Escola Regional de Alto Desempenho (ERAD-RS), pp. 4, Sociedade Brasileira de Computação (SBC), Três de Maio, BR, 2019. @inproceedings{MARON:ERAD:19,
title = {Benchmark Paramétrico para o Domínio do Paralelismo de Stream: Um Estudo de Caso com o Ferret da Suíte PARSEC},
author = {Carlos A. F. Maron and Dalvan Griebler and Luiz G. Fernandes},
url = {https://gmap.pucrs.br/dalvan/papers/2019/CR_ERAD_PG_Maron_2019.pdf},
year = {2019},
date = {2019-04-01},
booktitle = {Escola Regional de Alto Desempenho (ERAD-RS)},
pages = {4},
publisher = {Sociedade Brasileira de Computação (SBC)},
address = {Três de Maio, BR},
abstract = {Benchmarks são aplicações sintéticas que servem para avaliar e com-parar o desempenho de sistemas computacionais. Torná-los parametrizáveispode gerar condições diferenciadas de execuções. Porém, a técnica é pouco ex-plorada nos tradicionais e atuais benchmarks. Portanto, esse trabalho avalia oimpacto da parametrização de características do domínio de stream no Ferret.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Benchmarks são aplicações sintéticas que servem para avaliar e com-parar o desempenho de sistemas computacionais. Torná-los parametrizáveispode gerar condições diferenciadas de execuções. Porém, a técnica é pouco ex-plorada nos tradicionais e atuais benchmarks. Portanto, esse trabalho avalia oimpacto da parametrização de características do domínio de stream no Ferret. |
| Rista, Cassiano; Griebler, Dalvan; Fernandes, Luiz G. Proposta de Grau de Paralelismo Autoadaptativo com MPI-2 para a DSL SPar Inproceedings In: Escola Regional de Alto Desempenho (ERAD-RS), pp. 4, Sociedade Brasileira de Computação (SBC), Três de Maio, BR, 2019. @inproceedings{RISTA:ERAD:19,
title = {Proposta de Grau de Paralelismo Autoadaptativo com MPI-2 para a DSL SPar},
author = {Cassiano Rista and Dalvan Griebler and Luiz G. Fernandes},
url = {https://gmap.pucrs.br/dalvan/papers/2019/CR_ERAD_PG_Rista_2019.pdf},
year = {2019},
date = {2019-04-01},
booktitle = {Escola Regional de Alto Desempenho (ERAD-RS)},
pages = {4},
publisher = {Sociedade Brasileira de Computação (SBC)},
address = {Três de Maio, BR},
abstract = {Este artigo apresenta o projeto de um módulo autoadaptativo paracontrole do grau de paralelismo à ser integrado a DSL SPar. O módulo paraaplicações paralelas distribuídas de stream permite a criação de processos emtempo de execução, seleção da política de escalonamento, balanceamento decarga, ordenamento e serialização, adaptando o grau de paralelismo de formaautônoma sem a necessidade de definição de thresholds por parte do usuário.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Este artigo apresenta o projeto de um módulo autoadaptativo paracontrole do grau de paralelismo à ser integrado a DSL SPar. O módulo paraaplicações paralelas distribuídas de stream permite a criação de processos emtempo de execução, seleção da política de escalonamento, balanceamento decarga, ordenamento e serialização, adaptando o grau de paralelismo de formaautônoma sem a necessidade de definição de thresholds por parte do usuário. |
| Stein, Charles M.; Rockenbach, Dinei A.; Griebler, Dalvan Paralelização do Dedup para Sistemas Multi-core com GPUs Inproceedings In: 19th Escola Regional de Alto Desempenho da Região Sul (ERAD/RS), Sociedade Brasileira de Computação, Três de Maio, RS, Brazil, 2019. @inproceedings{larcc:paralelizacao_multicore_GPU:ERAD:19,
title = {Paralelização do Dedup para Sistemas Multi-core com GPUs},
author = {Charles M. Stein and Dinei A. Rockenbach and Dalvan Griebler},
url = {http://larcc.setrem.com.br/wp-content/uploads/2019/04/192087.pdf},
year = {2019},
date = {2019-04-01},
booktitle = {19th Escola Regional de Alto Desempenho da Região Sul (ERAD/RS)},
publisher = {Sociedade Brasileira de Computação},
address = {Três de Maio, RS, Brazil},
abstract = {O maior volume de dados gerado, trafegado e processado aumentaa demanda por mais poder de processamento e por algoritmos de compressãoeficientes. Este trabalho tem como objetivo explorar o paralelismo de streampara arquiteturas multi-core com GPUs na aplicação Dedup, usando SPar comCUDA e OpenCL. Apesar do desempenho não ser o esperado, o artigo contribuicom uma análise detalhada dos resultados e sugestões futuras de melhorias.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
O maior volume de dados gerado, trafegado e processado aumentaa demanda por mais poder de processamento e por algoritmos de compressãoeficientes. Este trabalho tem como objetivo explorar o paralelismo de streampara arquiteturas multi-core com GPUs na aplicação Dedup, usando SPar comCUDA e OpenCL. Apesar do desempenho não ser o esperado, o artigo contribuicom uma análise detalhada dos resultados e sugestões futuras de melhorias. |
| Stein, Charles M.; Stein, Joao V.; Boz, Leonardo; Rockenbach, Dinei A.; Griebler, Dalvan Mandelbrot Streaming para Sistemas Multi-core com GPUs Inproceedings In: 19th Escola Regional de Alto Desempenho da Região Sul (ERAD/RS), Sociedade Brasileira de Computação, Três de Maio, RS, Brazil, 2019. @inproceedings{larcc:mandelbrot_multicore_GPU:ERAD:19,
title = {Mandelbrot Streaming para Sistemas Multi-core com GPUs},
author = {Charles M. Stein and Joao V. Stein and Leonardo Boz and Dinei A. Rockenbach and Dalvan Griebler},
url = {http://larcc.setrem.com.br/wp-content/uploads/2019/04/192109.pdf},
year = {2019},
date = {2019-04-01},
booktitle = {19th Escola Regional de Alto Desempenho da Região Sul (ERAD/RS)},
publisher = {Sociedade Brasileira de Computação},
address = {Três de Maio, RS, Brazil},
abstract = {Este trabalho visa explorar o paralelismo na aplicação MandelbrotStreamingpara arquiteturas multi-core com GPUs, usando as bibliotecas Fast-Flow, TBB e SPar com CUDA. A implementação do paralelismo foi baseada nopadrão farm, alcançando speedup de 16x no sistema multi-core e de 77x em umambiente multi-core com duas GPUs. Os resultados evidenciam um melhor de-sempenho no uso de GPUs embora tenham sido identificadas futuras melhorias.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Este trabalho visa explorar o paralelismo na aplicação MandelbrotStreamingpara arquiteturas multi-core com GPUs, usando as bibliotecas Fast-Flow, TBB e SPar com CUDA. A implementação do paralelismo foi baseada nopadrão farm, alcançando speedup de 16x no sistema multi-core e de 77x em umambiente multi-core com duas GPUs. Os resultados evidenciam um melhor de-sempenho no uso de GPUs embora tenham sido identificadas futuras melhorias. |
| Araujo, Gabriell A.; Hoffmann, Renato B.; Griebler, Dalvan; Fernandes, Luiz G. Avaliando o Paralelismo de Stream com Pthreads, OpenMP e SPar em Aplicações de Vídeo Inproceedings In: Escola Regional de Alto Desempenho (ERAD-RS), pp. 4, Sociedade Brasileira de Computação (SBC), Três de Maio, BR, 2019. @inproceedings{ARAUJO:stream:ERAD:19,
title = {Avaliando o Paralelismo de Stream com Pthreads, OpenMP e SPar em Aplicações de Vídeo},
author = {Gabriell A. Araujo and Renato B. Hoffmann and Dalvan Griebler and Luiz G. Fernandes},
url = {https://gmap.pucrs.br/dalvan/papers/2019/CR_ERAD_IC_Araujo_2019.pdf},
year = {2019},
date = {2019-04-01},
booktitle = {Escola Regional de Alto Desempenho (ERAD-RS)},
pages = {4},
publisher = {Sociedade Brasileira de Computação (SBC)},
address = {Três de Maio, BR},
abstract = {Visando estender os estudos de avaliação daSPar, efetuamos umaanálise comparativa entre SPar,Pthreadse OpenMP em aplicações destream. Os resultados revelam que o desempenho do código paralelo geradopela SPar se equipara com as implementações robustas nas consolidadas bibliotecas Pthreads e OpenMP. Não obstante, também encontramos pontos depossíveis melhorias na SPar.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Visando estender os estudos de avaliação daSPar, efetuamos umaanálise comparativa entre SPar,Pthreadse OpenMP em aplicações destream. Os resultados revelam que o desempenho do código paralelo geradopela SPar se equipara com as implementações robustas nas consolidadas bibliotecas Pthreads e OpenMP. Não obstante, também encontramos pontos depossíveis melhorias na SPar. |
| Araujo, Gabriell A.; Griebler, Dalvan; Fernandes, Luiz G. Revisando a Programação Paralela com CUDA nos Benchmarks EP e FT Inproceedings In: Escola Regional de Alto Desempenho (ERAD-RS), pp. 4, Sociedade Brasileira de Computação (SBC), Três de Maio, BR, 2019. @inproceedings{ARAUJO:gpu:ERAD:19,
title = {Revisando a Programação Paralela com CUDA nos Benchmarks EP e FT},
author = {Gabriell A. Araujo and Dalvan Griebler and Luiz G. Fernandes},
url = {https://gmap.pucrs.br/dalvan/papers/2019/CR_ERAD_IC_Gabriell_2019.pdf},
year = {2019},
date = {2019-04-01},
booktitle = {Escola Regional de Alto Desempenho (ERAD-RS)},
pages = {4},
publisher = {Sociedade Brasileira de Computação (SBC)},
address = {Três de Maio, BR},
abstract = {Este trabalho visa estender os estudos sobre o NAS Parallel Ben-chmarks (NPB), os quais possuem lacunas relevantes no contexto de GPUs.Os principais trabalhos da literatura consistem em implementações antigas,abrindo margens para possíveis questionamentos. Nessa direção, foram rea-lizados novos estudos de paralelização para GPUs das aplicações EP e FT. Osresultados foram similares ou melhores que o estado-da-arte.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Este trabalho visa estender os estudos sobre o NAS Parallel Ben-chmarks (NPB), os quais possuem lacunas relevantes no contexto de GPUs.Os principais trabalhos da literatura consistem em implementações antigas,abrindo margens para possíveis questionamentos. Nessa direção, foram rea-lizados novos estudos de paralelização para GPUs das aplicações EP e FT. Osresultados foram similares ou melhores que o estado-da-arte. |
| Justo, Gabriel B.; Vogel, Adriano; Griebler, Dalvan; Fernandes, Luiz G. Acelerando o Reconhecimento de Pessoas em Vídeos com MPI Inproceedings In: Escola Regional de Alto Desempenho (ERAD-RS), pp. 4, Sociedade Brasileira de Computação (SBC), Três de Maio, BR, 2019. @inproceedings{JUSTO:ERAD:19,
title = {Acelerando o Reconhecimento de Pessoas em Vídeos com MPI},
author = {Gabriel B. Justo and Adriano Vogel and Dalvan Griebler and Luiz G. Fernandes},
url = {https://gmap.pucrs.br/dalvan/papers/2019/CR_ERAD_IC_Justo_2019.pdf},
year = {2019},
date = {2019-04-01},
booktitle = {Escola Regional de Alto Desempenho (ERAD-RS)},
pages = {4},
publisher = {Sociedade Brasileira de Computação (SBC)},
address = {Três de Maio, BR},
abstract = {Diversas aplicações de processamento de vídeo demandam paralelismo para aumentar o desempenho. O objetivo deste trabalho é implementar etestar versões com processamento distribuído em aplicações de reconhecimentofacial em vídeos. As implementações foram avaliadas quanto ao seu desempe-nho. Os resultados mostraram que essas aplicações podem ter uma aceleraçãosignificativa em ambientes distribuídos.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Diversas aplicações de processamento de vídeo demandam paralelismo para aumentar o desempenho. O objetivo deste trabalho é implementar etestar versões com processamento distribuído em aplicações de reconhecimentofacial em vídeos. As implementações foram avaliadas quanto ao seu desempe-nho. Os resultados mostraram que essas aplicações podem ter uma aceleraçãosignificativa em ambientes distribuídos. |
| Maliszewski, Anderson M.; Fim, Gabriel R.; Maron, Carlos A. F.; Vogel, Adriano; Griebler, Dalvan Avaliação de Desempenho em Contêineres LXD com Aplicações Científicas na Nuvem OpenNebula Inproceedings In: 19th Escola Regional de Alto Desempenho da Região Sul (ERAD/RS), Sociedade Brasileira de Computação, Três de Maio, RS, Brazil, 2019. @inproceedings{larcc:desempenho_LXD_Opennebula:ERAD:19,
title = {Avaliação de Desempenho em Contêineres LXD com Aplicações Científicas na Nuvem OpenNebula},
author = {Anderson M. Maliszewski and Gabriel R. Fim and Carlos A. F. Maron and Adriano Vogel and Dalvan Griebler},
url = {http://larcc.setrem.com.br/wp-content/uploads/2019/04/192099.pdf},
year = {2019},
date = {2019-04-01},
booktitle = {19th Escola Regional de Alto Desempenho da Região Sul (ERAD/RS)},
publisher = {Sociedade Brasileira de Computação},
address = {Três de Maio, RS, Brazil},
abstract = {As nuvens privadas IaaS podem fornecer um ambiente atrativo paraaplicações científicas. No entanto, como existem diversos modelos de implan-tação e configuração, avaliar o desempenho dessas aplicações é um desafio.Este artigo tem como objetivo avaliar o desempenho de contêineres LXD ge-renciados pelo OpenNebula, utilizando os benchmarks da suite NPB-MPI. Osresultados mostram que o LXD não induz a grandes overheads no desempenho},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
As nuvens privadas IaaS podem fornecer um ambiente atrativo paraaplicações científicas. No entanto, como existem diversos modelos de implan-tação e configuração, avaliar o desempenho dessas aplicações é um desafio.Este artigo tem como objetivo avaliar o desempenho de contêineres LXD ge-renciados pelo OpenNebula, utilizando os benchmarks da suite NPB-MPI. Osresultados mostram que o LXD não induz a grandes overheads no desempenho |
| Stein, Charles Michael; Griebler, Dalvan; Danelutto, Marco; Fernandes, Luiz Gustavo Stream Parallelism on the LZSS Data Compression Application for Multi-Cores with GPUs Inproceedings doi In: 27th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP), pp. 247-251, IEEE, Pavia, Italy, 2019. @inproceedings{STEIN:LZSS-multigpu:PDP:19,
title = {Stream Parallelism on the LZSS Data Compression Application for Multi-Cores with GPUs},
author = {Charles Michael Stein and Dalvan Griebler and Marco Danelutto and Luiz Gustavo Fernandes},
url = {https://doi.org/10.1109/EMPDP.2019.8671624},
doi = {10.1109/EMPDP.2019.8671624},
year = {2019},
date = {2019-02-01},
booktitle = {27th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP)},
pages = {247-251},
publisher = {IEEE},
address = {Pavia, Italy},
series = {PDP'19},
abstract = {GPUs have been used to accelerate different data parallel applications. The challenge consists in using GPUs to accelerate stream processing applications. Our goal is to investigate and evaluate whether stream parallel applications may benefit from parallel execution on both CPU and GPU cores. In this paper, we introduce new parallel algorithms for the Lempel-Ziv-Storer-Szymanski (LZSS) data compression application. We implemented the algorithms targeting both CPUs and GPUs. GPUs have been used with CUDA and OpenCL to exploit inner algorithm data parallelism. Outer stream parallelism has been exploited using CPU cores through SPar. The parallel implementation of LZSS achieved 135 fold speedup using a multi-core CPU and two GPUs. We also observed speedups in applications where we were not expecting to get it using the same combine data-stream parallel exploitation techniques.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
GPUs have been used to accelerate different data parallel applications. The challenge consists in using GPUs to accelerate stream processing applications. Our goal is to investigate and evaluate whether stream parallel applications may benefit from parallel execution on both CPU and GPU cores. In this paper, we introduce new parallel algorithms for the Lempel-Ziv-Storer-Szymanski (LZSS) data compression application. We implemented the algorithms targeting both CPUs and GPUs. GPUs have been used with CUDA and OpenCL to exploit inner algorithm data parallelism. Outer stream parallelism has been exploited using CPU cores through SPar. The parallel implementation of LZSS achieved 135 fold speedup using a multi-core CPU and two GPUs. We also observed speedups in applications where we were not expecting to get it using the same combine data-stream parallel exploitation techniques. |
| Maron, Carlos A. F.; Vogel, Adriano; Griebler, Dalvan; Fernandes, Luiz Gustavo Should PARSEC Benchmarks be More Parametric? A Case Study with Dedup Inproceedings doi In: 27th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP), pp. 217-221, IEEE, Pavia, Italy, 2019. @inproceedings{MARON:parametric-parsec:PDP:19,
title = {Should PARSEC Benchmarks be More Parametric? A Case Study with Dedup},
author = {Carlos A. F. Maron and Adriano Vogel and Dalvan Griebler and Luiz Gustavo Fernandes},
url = {https://doi.org/10.1109/EMPDP.2019.8671592},
doi = {10.1109/EMPDP.2019.8671592},
year = {2019},
date = {2019-02-01},
booktitle = {27th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP)},
pages = {217-221},
publisher = {IEEE},
address = {Pavia, Italy},
series = {PDP'19},
abstract = {Parallel applications of the same domain can present similar patterns of behavior and characteristics. Characterizing common application behaviors can help for understanding performance aspects in the real-world scenario. One way to better understand and evaluate applications' characteristics is by using customizable/parametric benchmarks that enable users to represent important characteristics at run-time. We observed that parameterization techniques should be better exploited in the available benchmarks, especially on stream processing domain. For instance, although widely used, the stream processing benchmarks available in PARSEC do not support the simulation and evaluation of relevant and modern characteristics. Therefore, our goal is to identify the stream parallelism characteristics present in PARSEC. We also implemented a ready to use parameterization support and evaluated the application behaviors considering relevant performance metrics for stream parallelism (service time, throughput, latency). We choose Dedup to be our case study. The experimental results have shown performance improvements in our parameterization support for Dedup. Moreover, this support increased the customization space for benchmark users, which is simple to use. In the future, our solution can be potentially explored on different parallel architectures and parallel programming frameworks.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Parallel applications of the same domain can present similar patterns of behavior and characteristics. Characterizing common application behaviors can help for understanding performance aspects in the real-world scenario. One way to better understand and evaluate applications' characteristics is by using customizable/parametric benchmarks that enable users to represent important characteristics at run-time. We observed that parameterization techniques should be better exploited in the available benchmarks, especially on stream processing domain. For instance, although widely used, the stream processing benchmarks available in PARSEC do not support the simulation and evaluation of relevant and modern characteristics. Therefore, our goal is to identify the stream parallelism characteristics present in PARSEC. We also implemented a ready to use parameterization support and evaluated the application behaviors considering relevant performance metrics for stream parallelism (service time, throughput, latency). We choose Dedup to be our case study. The experimental results have shown performance improvements in our parameterization support for Dedup. Moreover, this support increased the customization space for benchmark users, which is simple to use. In the future, our solution can be potentially explored on different parallel architectures and parallel programming frameworks. |
| Serpa, Matheus S.; Moreira, Francis B.; Navaux, Philippe O. A.; Cruz, Eduardo H. M.; Diener, Matthias; Griebler, Dalvan; Fernandes, Luiz Gustavo Memory Performance and Bottlenecks in Multicore and GPU Architectures Inproceedings doi In: 27th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP), pp. 233-236, IEEE, Pavia, Italy, 2019. @inproceedings{SERPA:memory-gpu-multicore:PDP:19,
title = {Memory Performance and Bottlenecks in Multicore and GPU Architectures},
author = {Matheus S. Serpa and Francis B. Moreira and Philippe O. A. Navaux and Eduardo H. M. Cruz and Matthias Diener and Dalvan Griebler and Luiz Gustavo Fernandes},
url = {https://doi.org/10.1109/EMPDP.2019.8671628},
doi = {10.1109/EMPDP.2019.8671628},
year = {2019},
date = {2019-02-01},
booktitle = {27th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP)},
pages = {233-236},
publisher = {IEEE},
address = {Pavia, Italy},
series = {PDP'19},
abstract = {Nowadays, there are several different architectures available not only for the industry, but also for normal consumers. Traditional multicore processors, GPUs, accelerators such as the Sunway SW26010, or even energy efficiency-driven processors such as the ARM family, present very different architectural characteristics. This wide range of characteristics presents a challenge for the developers of applications. Developers must deal with different instruction sets, memory hierarchies, or even different programming paradigms when programming for these architectures. Therefore, the same application can perform well when executing on one architecture, but poorly on another architecture. To optimize an application, it is important to have a deep understanding of how it behaves on different architectures. The related work in this area mostly focuses on a limited analysis encompassing execution time and energy. In this paper, we perform a detailed investigation on the impact of the memory subsystem of different architectures, which is one of the most important aspects to be considered. For this study, we performed experiments in the Broadwell CPU and Pascal GPU, using applications from the Rodinia benchmark suite. In this way, we were able to understand why an application performs well on one architecture and poorly on others.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Nowadays, there are several different architectures available not only for the industry, but also for normal consumers. Traditional multicore processors, GPUs, accelerators such as the Sunway SW26010, or even energy efficiency-driven processors such as the ARM family, present very different architectural characteristics. This wide range of characteristics presents a challenge for the developers of applications. Developers must deal with different instruction sets, memory hierarchies, or even different programming paradigms when programming for these architectures. Therefore, the same application can perform well when executing on one architecture, but poorly on another architecture. To optimize an application, it is important to have a deep understanding of how it behaves on different architectures. The related work in this area mostly focuses on a limited analysis encompassing execution time and energy. In this paper, we perform a detailed investigation on the impact of the memory subsystem of different architectures, which is one of the most important aspects to be considered. For this study, we performed experiments in the Broadwell CPU and Pascal GPU, using applications from the Rodinia benchmark suite. In this way, we were able to understand why an application performs well on one architecture and poorly on others. |
2018
|
| Ewald, Endrius; Vogel, Adriano; Rista, Cassiano; Griebler, Dalvan; Manssour, Isabel; Fernandes, Luiz G. Parallel and Distributed Processing Support for a Geospatial Data Visualization DSL Inproceedings doi In: Symposium on High Performance Computing Systems (WSCAD), pp. 221-228, IEEE, São Paulo, Brazil, 2018. @inproceedings{EWALD:WSCAD:18,
title = {Parallel and Distributed Processing Support for a Geospatial Data Visualization DSL},
author = {Endrius Ewald and Adriano Vogel and Cassiano Rista and Dalvan Griebler and Isabel Manssour and Luiz G. Fernandes},
url = {https://doi.org/10.1109/WSCAD.2018.00042},
doi = {10.1109/WSCAD.2018.00042},
year = {2018},
date = {2018-10-01},
booktitle = {Symposium on High Performance Computing Systems (WSCAD)},
pages = {221-228},
publisher = {IEEE},
address = {São Paulo, Brazil},
abstract = {The amount of data generated worldwide related to geolocalization has exponentially increased. However, the fast processing of this amount of data is a challenge from the programming perspective, and many available solutions require learning a variety of tools and programming languages. This paper introduces the support for parallel and distributed processing in a DSL for Geospatial Data Visualization to speed up the data pre-processing phase. The results have shown the MPI version with dynamic data distribution performing better under medium and large data set files, while MPI-I/O version achieved the best performance with small data set files.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The amount of data generated worldwide related to geolocalization has exponentially increased. However, the fast processing of this amount of data is a challenge from the programming perspective, and many available solutions require learning a variety of tools and programming languages. This paper introduces the support for parallel and distributed processing in a DSL for Geospatial Data Visualization to speed up the data pre-processing phase. The results have shown the MPI version with dynamic data distribution performing better under medium and large data set files, while MPI-I/O version achieved the best performance with small data set files. |
| Maliszewski, Anderson M; Griebler, Dalvan; Vogel, Adriano; Schepke, Claudio On the Performance of Multithreading Applications under Private Cloud Conditions Inproceedings doi In: Symposium on High Performance Computing Systems (WSCAD), pp. 273-273, IEEE, São Paulo, Brazil, 2018. @inproceedings{larcc:multithreading_cloud:WSCAD:18,
title = {On the Performance of Multithreading Applications under Private Cloud Conditions},
author = {Anderson M Maliszewski and Dalvan Griebler and Adriano Vogel and Claudio Schepke},
url = {https://doi.org/10.1109/WSCAD.2018.00055},
doi = {10.1109/WSCAD.2018.00055},
year = {2018},
date = {2018-10-01},
booktitle = {Symposium on High Performance Computing Systems (WSCAD)},
pages = {273-273},
publisher = {IEEE},
address = {São Paulo, Brazil},
abstract = {IaaS private clouds provide an attractive environment for scientific applications. However, the performance is a challenge, as additional abstraction layers imposed by the virtualization can cause overheads and bottlenecks. This paper contributes to a performance analysis of applications with dedicated and shared resources environments under private cloud conditions, deployed with container (LXC) or kernel-based (KVM) instances. We selected five benchmarks from PARSEC suite. In the experimental results, identify a performance pattern of behavior among the applications was hard. For a set of multi-threading applications, the KVM-based cloud instances achieved better performance, however, in the other set of applications, the LXC-based cloud instances performed better.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
IaaS private clouds provide an attractive environment for scientific applications. However, the performance is a challenge, as additional abstraction layers imposed by the virtualization can cause overheads and bottlenecks. This paper contributes to a performance analysis of applications with dedicated and shared resources environments under private cloud conditions, deployed with container (LXC) or kernel-based (KVM) instances. We selected five benchmarks from PARSEC suite. In the experimental results, identify a performance pattern of behavior among the applications was hard. For a set of multi-threading applications, the KVM-based cloud instances achieved better performance, however, in the other set of applications, the LXC-based cloud instances performed better. |
| Griebler, Dalvan; Sensi, Daniele De; Vogel, Adriano; Danelutto, Marco; Fernandes, Luiz Gustavo Service Level Objectives via C++11 Attributes Inproceedings doi In: Euro-Par 2018: Parallel Processing Workshops, pp. 745-756, Springer, Turin, Italy, 2018. @inproceedings{GRIEBLER:SLO-SPar-Nornir:REPARA:18,
title = {Service Level Objectives via C++11 Attributes},
author = {Dalvan Griebler and Daniele De Sensi and Adriano Vogel and Marco Danelutto and Luiz Gustavo Fernandes},
url = {http://dx.doi.org/10.1007/978-3-030-10549-5_58},
doi = {10.1007/978-3-030-10549-5_58},
year = {2018},
date = {2018-08-01},
booktitle = {Euro-Par 2018: Parallel Processing Workshops},
pages = {745-756},
publisher = {Springer},
address = {Turin, Italy},
series = {Lecture Notes in Computer Science},
abstract = {In recent years, increasing attention has been given to the possibility of guaranteeing Service Level Objectives (SLOs) to users about their applications, either regarding performance or power consumption. SLO can be implemented for parallel applications since they can provide many control knobs (e.g., the number of threads to use, the clock frequency of the cores, etc.) to tune the performance and power consumption of the application. Different from most of the existing approaches, we target sequential stream processing applications by proposing a solution based on C++ annotations. The user specifies which parts of the code to parallelize and what type of requirements should be enforced on that part of the code. Our solution first automatically parallelizes the annotated code and then applies self-adaptation approaches at run-time to enforce the user-expressed objectives. We ran experiments on different real-world applications, showing its simplicity and effectiveness.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
In recent years, increasing attention has been given to the possibility of guaranteeing Service Level Objectives (SLOs) to users about their applications, either regarding performance or power consumption. SLO can be implemented for parallel applications since they can provide many control knobs (e.g., the number of threads to use, the clock frequency of the cores, etc.) to tune the performance and power consumption of the application. Different from most of the existing approaches, we target sequential stream processing applications by proposing a solution based on C++ annotations. The user specifies which parts of the code to parallelize and what type of requirements should be enforced on that part of the code. Our solution first automatically parallelizes the annotated code and then applies self-adaptation approaches at run-time to enforce the user-expressed objectives. We ran experiments on different real-world applications, showing its simplicity and effectiveness. |
| Vogel, Adriano; Griebler, Dalvan; Sensi, Daniele De; Danelutto, Marco; Fernandes, Luiz Gustavo Autonomic and Latency-Aware Degree of Parallelism Management in SPar Inproceedings doi In: Euro-Par 2018: Parallel Processing Workshops, pp. 28-39, Springer, Turin, Italy, 2018. @inproceedings{VOGEL:Adaptive-Latency-SPar:AutoDaSP:18,
title = {Autonomic and Latency-Aware Degree of Parallelism Management in SPar},
author = {Adriano Vogel and Dalvan Griebler and Daniele De Sensi and Marco Danelutto and Luiz Gustavo Fernandes},
url = {http://dx.doi.org/10.1007/978-3-030-10549-5_3},
doi = {10.1007/978-3-030-10549-5_3},
year = {2018},
date = {2018-08-01},
booktitle = {Euro-Par 2018: Parallel Processing Workshops},
pages = {28-39},
publisher = {Springer},
address = {Turin, Italy},
series = {Lecture Notes in Computer Science},
abstract = {Stream processing applications became a representative workload in current computing systems. A significant part of these applications demands parallelism to increase performance. However, programmers are often facing a trade-off between coding productivity and performance when introducing parallelism. SPar was created for balancing this trade-off to the application programmers by using the C++11 attributes’ annotation mechanism. In SPar and other programming frameworks for stream processing applications, the manual definition of the number of replicas to be used for the stream operators is a challenge. In addition to that, low latency is required by several stream processing applications. We noted that explicit latency requirements are poorly considered on the state-of-the-art parallel programming frameworks. Since there is a direct relationship between the number of replicas and the latency of the application, in this work we propose an autonomic and adaptive strategy to choose the proper number of replicas in SPar to address latency constraints. We experimentally evaluated our implemented strategy and demonstrated its effectiveness on a real-world application, demonstrating that our adaptive strategy can provide higher abstraction levels while automatically managing the latency.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Stream processing applications became a representative workload in current computing systems. A significant part of these applications demands parallelism to increase performance. However, programmers are often facing a trade-off between coding productivity and performance when introducing parallelism. SPar was created for balancing this trade-off to the application programmers by using the C++11 attributes’ annotation mechanism. In SPar and other programming frameworks for stream processing applications, the manual definition of the number of replicas to be used for the stream operators is a challenge. In addition to that, low latency is required by several stream processing applications. We noted that explicit latency requirements are poorly considered on the state-of-the-art parallel programming frameworks. Since there is a direct relationship between the number of replicas and the latency of the application, in this work we propose an autonomic and adaptive strategy to choose the proper number of replicas in SPar to address latency constraints. We experimentally evaluated our implemented strategy and demonstrated its effectiveness on a real-world application, demonstrating that our adaptive strategy can provide higher abstraction levels while automatically managing the latency. |
 | Griebler, Dalvan; Hoffmann, Renato B.; Danelutto, Marco; Fernandes, Luiz Gustavo Stream Parallelism with Ordered Data Constraints on Multi-Core Systems Journal Article doi In: The Journal of Supercomputing, vol. 75, no. 8, pp. 4042-4061, 2018, ISSN: 0920-8542. @article{GRIEBLER:JS:18,
title = {Stream Parallelism with Ordered Data Constraints on Multi-Core Systems},
author = {Dalvan Griebler and Renato B. Hoffmann and Marco Danelutto and Luiz Gustavo Fernandes},
url = {https://doi.org/10.1007/s11227-018-2482-7},
doi = {10.1007/s11227-018-2482-7},
issn = {0920-8542},
year = {2018},
date = {2018-07-01},
urldate = {2018-07-01},
journal = {The Journal of Supercomputing},
volume = {75},
number = {8},
pages = {4042-4061},
publisher = {Springer},
abstract = {It is often a challenge to keep input/output tasks/results in order for parallel computations ver data streams, particularly when stateless task operators are replicated to increase parallelism when there are irregular tasks. Maintaining input/output order requires additional coding effort and may significantly impact the application's actual throughput. Thus, we propose a new implementation technique designed to be easily integrated with any of the existing C++ parallel programming frameworks that support stream parallelism. In this paper, it is first implemented and studied using SPar, our high-level domain-specific language for stream parallelism. We discuss the results of a set of experiments with real-world applications revealing how significant performance improvements may be achieved when our proposed solution is integrated within SPar, especially for data compression applications. Also, we show the results of experiments performed after integrating our solution within FastFlow and TBB, revealing no significant overheads.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
It is often a challenge to keep input/output tasks/results in order for parallel computations ver data streams, particularly when stateless task operators are replicated to increase parallelism when there are irregular tasks. Maintaining input/output order requires additional coding effort and may significantly impact the application's actual throughput. Thus, we propose a new implementation technique designed to be easily integrated with any of the existing C++ parallel programming frameworks that support stream parallelism. In this paper, it is first implemented and studied using SPar, our high-level domain-specific language for stream parallelism. We discuss the results of a set of experiments with real-world applications revealing how significant performance improvements may be achieved when our proposed solution is integrated within SPar, especially for data compression applications. Also, we show the results of experiments performed after integrating our solution within FastFlow and TBB, revealing no significant overheads. |
| Maliszewski, Anderson M; Griebler, Dalvan; Schepke, Claudio; Ditter, Alexander; Fey, Dietmar; Fernandes, Luiz Gustavo The NAS Benchmark Kernels for Single and Multi-Tenant Cloud Instances with LXC/KVM Inproceedings doi In: International Conference on High Performance Computing & Simulation (HPCS), pp. 359-366, IEEE, Orleans, France, 2018. @inproceedings{larcc:NAS_cloud_LXC_KVM:HPCS:2018,
title = {The NAS Benchmark Kernels for Single and Multi-Tenant Cloud Instances with LXC/KVM},
author = {Anderson M Maliszewski and Dalvan Griebler and Claudio Schepke and Alexander Ditter and Dietmar Fey and Luiz Gustavo Fernandes},
url = {https://doi.org/10.1109/HPCS.2018.00066},
doi = {10.1109/HPCS.2018.00066},
year = {2018},
date = {2018-07-01},
booktitle = {International Conference on High Performance Computing & Simulation (HPCS)},
pages = {359-366},
publisher = {IEEE},
address = {Orleans, France},
series = {HPCS'18},
abstract = {Private IaaS clouds are an attractive environment for scientific workloads and applications. It provides advantages such as almost instantaneous availability of high-performance computing in a single node as well as compute clusters, easy access for researchers, and users that do not have access to conventional supercomputers. Furthermore, a cloud infrastructure provides elasticity and scalability to ensure and manage any software dependency on the system with no third-party dependency for researchers. However, one of the biggest challenges is to avoid significant performance degradation when migrating these applications from physical nodes to a cloud environment. Also, we lack more research investigations for multi-tenant cloud instances. In this paper, our goal is to perform a comparative performance evaluation of scientific applications with single and multi-tenancy cloud instances using KVM and LXC virtualization technologies under private cloud conditions. All analyses and evaluations were carried out based on NAS Benchmark kernels to simulate different types of workloads. We applied statistic significance tests to highlight the differences. The results have shown that applications running on LXC-based cloud instances outperform KVM-based cloud instances in 93.75% of the experiments w.r.t single tenant. Regarding multi-tenant, LXC instances outperform KVM instances in 45% of the results, where the performance differences were not as significant as expected.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Private IaaS clouds are an attractive environment for scientific workloads and applications. It provides advantages such as almost instantaneous availability of high-performance computing in a single node as well as compute clusters, easy access for researchers, and users that do not have access to conventional supercomputers. Furthermore, a cloud infrastructure provides elasticity and scalability to ensure and manage any software dependency on the system with no third-party dependency for researchers. However, one of the biggest challenges is to avoid significant performance degradation when migrating these applications from physical nodes to a cloud environment. Also, we lack more research investigations for multi-tenant cloud instances. In this paper, our goal is to perform a comparative performance evaluation of scientific applications with single and multi-tenancy cloud instances using KVM and LXC virtualization technologies under private cloud conditions. All analyses and evaluations were carried out based on NAS Benchmark kernels to simulate different types of workloads. We applied statistic significance tests to highlight the differences. The results have shown that applications running on LXC-based cloud instances outperform KVM-based cloud instances in 93.75% of the experiments w.r.t single tenant. Regarding multi-tenant, LXC instances outperform KVM instances in 45% of the results, where the performance differences were not as significant as expected. |
| Klein, Maikel; Maliszewski, Anderson Mattheus; Griebler, Dalvan Avaliação do Desempenho do Protocolo Bonding em Máquinas Virtuais LXC e KVM Inproceedings In: 15th Escola Regional de Redes de Computadores (ERRC), pp. 1-8, Sociedade Brasileira de Computação, Pelotas, BR, 2018. @inproceedings{larcc:link_agreggation:ERRC:18,
title = {Avaliação do Desempenho do Protocolo Bonding em Máquinas Virtuais LXC e KVM},
author = {Maikel Klein and Anderson Mattheus Maliszewski and Dalvan Griebler},
url = {http://larcc.setrem.com.br/wpcontent/uploads/2018/11/ERRC_2018__Link_Aggregation_.pdf},
year = {2018},
date = {2018-07-01},
booktitle = {15th Escola Regional de Redes de Computadores (ERRC)},
pages = {1-8},
publisher = {Sociedade Brasileira de Computação},
address = {Pelotas, BR},
abstract = {O processamento de grandes volumes de dados (Big Data) e seu armazenamento distribuído vem aumentado gradualmente o uso da rede. Com isso, torna-se necessário o uso de tecnologias para otimizar a largura de banda. Uma das soluções de baixo custo e fácil implementação é a agregação de link. Além disso, a virtualização, usada como base na computação em nuvem, oferece vários benefícios utilizados no Big Data. O objetivo deste trabalho é avaliar o desempenho de rede usando a agregação de link com o protocolo bonding em máquinas virtuais LXC e KVM. Os resultados mostram que o protocolo bonding tem comportamento similar com ambos tipos de virtualização.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
O processamento de grandes volumes de dados (Big Data) e seu armazenamento distribuído vem aumentado gradualmente o uso da rede. Com isso, torna-se necessário o uso de tecnologias para otimizar a largura de banda. Uma das soluções de baixo custo e fácil implementação é a agregação de link. Além disso, a virtualização, usada como base na computação em nuvem, oferece vários benefícios utilizados no Big Data. O objetivo deste trabalho é avaliar o desempenho de rede usando a agregação de link com o protocolo bonding em máquinas virtuais LXC e KVM. Os resultados mostram que o protocolo bonding tem comportamento similar com ambos tipos de virtualização. |
| Griebler, Dalvan; Vogel, Adriano; Maron, Carlos A F; Maliszewski, Anderson M; Schepke, Claudio; Fernandes, Luiz Gustavo Performance of Data Mining, Media, and Financial Applications under Private Cloud Conditions Inproceedings doi In: IEEE Symposium on Computers and Communications (ISCC), pp. 1530-1346, IEEE, Natal, Brazil, 2018. @inproceedings{larcc:parsec_cloudstack_lxc_kvm:ISCC:2018,
title = {Performance of Data Mining, Media, and Financial Applications under Private Cloud Conditions},
author = {Dalvan Griebler and Adriano Vogel and Carlos A F Maron and Anderson M Maliszewski and Claudio Schepke and Luiz Gustavo Fernandes},
url = {https://dx.doi.org/10.1109/ISCC.2018.8538759},
doi = {10.1109/ISCC.2018.8538759},
year = {2018},
date = {2018-06-01},
booktitle = {IEEE Symposium on Computers and Communications (ISCC)},
pages = {1530-1346},
publisher = {IEEE},
address = {Natal, Brazil},
series = {ISCC'18},
abstract = {This paper contributes to a performance analysis of real-world workloads under private cloud conditions. We selected six benchmarks from PARSEC related to three mainstream application domains (financial, data mining, and media processing). Our goal was to evaluate these application domains in different cloud instances and deployment environments, concerning container or kernel-based instances and using dedicated or shared machine resources. Experiments have shown that performance varies according to the application characteristics, virtualization technology, and cloud environment. Results highlighted that financial, data mining, and media processing applications running in the LXC instances tend to outperform KVM when there is a dedicated machine resource environment. However, when two instances are sharing the same machine resources, these applications tend to achieve better performance in the KVM instances. Finally, financial applications achieved better performance in the cloud than media and data mining.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper contributes to a performance analysis of real-world workloads under private cloud conditions. We selected six benchmarks from PARSEC related to three mainstream application domains (financial, data mining, and media processing). Our goal was to evaluate these application domains in different cloud instances and deployment environments, concerning container or kernel-based instances and using dedicated or shared machine resources. Experiments have shown that performance varies according to the application characteristics, virtualization technology, and cloud environment. Results highlighted that financial, data mining, and media processing applications running in the LXC instances tend to outperform KVM when there is a dedicated machine resource environment. However, when two instances are sharing the same machine resources, these applications tend to achieve better performance in the KVM instances. Finally, financial applications achieved better performance in the cloud than media and data mining. |
| Rista, Cassiano; Teixeira, Marcelo; Griebler, Dalvan; Fernandes, Luiz Gustavo Evaluating, Estimating, and Improving Network Performance in Container-based Clouds Inproceedings doi In: IEEE Symposium on Computers and Communications (ISCC), pp. 1530-1346, IEEE, Natal, Brazil, 2018. @inproceedings{larcc:network_performance_container:ISCC:2018,
title = {Evaluating, Estimating, and Improving Network Performance in Container-based Clouds},
author = {Cassiano Rista and Marcelo Teixeira and Dalvan Griebler and Luiz Gustavo Fernandes},
url = {https://doi.org/10.1109/ISCC.2018.8538558},
doi = {10.1109/ISCC.2018.8538558},
year = {2018},
date = {2018-06-01},
booktitle = {IEEE Symposium on Computers and Communications (ISCC)},
pages = {1530-1346},
publisher = {IEEE},
address = {Natal, Brazil},
series = {ISCC'18},
abstract = {Cloud computing has recently attracted a great deal of interest from both industry and academia, emerging as an important paradigm to improve resource utilization, efficiency, flexibility, and pay-per-use. However, cloud platforms inherently include a virtualization layer that imposes performance degradation on network-intensive applications. Thus, it is crucial to anticipate possible performance degradation to resolve system bottlenecks. This paper uses the Petri Nets approach to create different models for evaluating, estimating, and improving network performance in container-based cloud environments. Based on model estimations, we assessed the network bandwidth utilization of the system under different setups. Then, by identifying possible bottlenecks, we show how the system could be modified to improve performance. We then tested how the model would behave through real-world experiments. When the model indicates probable bandwidth saturation, we propose a link aggregation approach to increase bandwidth, using lightweight virtualization to reduce virtualization overhead. Results reveal that our model anticipates the structural and behavioral characteristics of the network in the cloud environment. Therefore, it systematically improves network efficiency, which saves effort, time, and money.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Cloud computing has recently attracted a great deal of interest from both industry and academia, emerging as an important paradigm to improve resource utilization, efficiency, flexibility, and pay-per-use. However, cloud platforms inherently include a virtualization layer that imposes performance degradation on network-intensive applications. Thus, it is crucial to anticipate possible performance degradation to resolve system bottlenecks. This paper uses the Petri Nets approach to create different models for evaluating, estimating, and improving network performance in container-based cloud environments. Based on model estimations, we assessed the network bandwidth utilization of the system under different setups. Then, by identifying possible bottlenecks, we show how the system could be modified to improve performance. We then tested how the model would behave through real-world experiments. When the model indicates probable bandwidth saturation, we propose a link aggregation approach to increase bandwidth, using lightweight virtualization to reduce virtualization overhead. Results reveal that our model anticipates the structural and behavioral characteristics of the network in the cloud environment. Therefore, it systematically improves network efficiency, which saves effort, time, and money. |
 | Rockenbach, Dinei A.; Anderle, Nadine; Griebler, Dalvan; Souza, Samuel Estudo Comparativo de Bancos de Dados NoSQL Journal Article doi In: Revista Eletrônica Argentina-Brasil de Tecnologias da Informação e da Comunicação (REABTIC), vol. 1, no. 8, 2018. @article{larcc:comparativo_nosql:REABTIC:18,
title = {Estudo Comparativo de Bancos de Dados NoSQL},
author = {Dinei A. Rockenbach and Nadine Anderle and Dalvan Griebler and Samuel Souza},
url = {https://revistas.setrem.com.br/index.php/reabtic/article/view/286},
doi = {10.5281/zenodo.1228503},
year = {2018},
date = {2018-04-01},
journal = {Revista Eletrônica Argentina-Brasil de Tecnologias da Informação e da Comunicação (REABTIC)},
volume = {1},
number = {8},
publisher = {SETREM},
address = {Três de Maio, RS, Brazil},
abstract = {NoSQL databases emerged to fill limitations of the relational databases. The many options for each one of the categories, and their distinct characteristics and focus makes this assessment very difficult for decision makers. Most of the time, decisions are taken without the attention and background deserved due to the related complexities. This article aims to compare the relevant characteristics of each database, abstracting the information that bases the market marketing of them. We concluded that although the databases are labeled in a specific category, there is a significant disparity in the functionalities offered by each of them. Also, we observed that new databases are emerging even though there are well-established databases in each one of the categories studied. Finally, it is very challenging to suggest the best database for each category because each scenario has its requirements, which requires a careful analysis where our work can help to simplify this kind of decision.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
NoSQL databases emerged to fill limitations of the relational databases. The many options for each one of the categories, and their distinct characteristics and focus makes this assessment very difficult for decision makers. Most of the time, decisions are taken without the attention and background deserved due to the related complexities. This article aims to compare the relevant characteristics of each database, abstracting the information that bases the market marketing of them. We concluded that although the databases are labeled in a specific category, there is a significant disparity in the functionalities offered by each of them. Also, we observed that new databases are emerging even though there are well-established databases in each one of the categories studied. Finally, it is very challenging to suggest the best database for each category because each scenario has its requirements, which requires a careful analysis where our work can help to simplify this kind of decision. |
| Stein, Charles M.; Griebler, Dalvan Explorando o Paralelismo de Stream em CPU e de Dados em GPU na Aplicação de Filtro Sobel Inproceedings In: 18th Escola Regional de Alto Desempenho do Estado do Rio Grande do Sul (ERAD/RS), pp. 137-140, Sociedade Brasileira de Computação, Porto Alegre, RS, Brazil, 2018. @inproceedings{larcc:stream_gpu_cuda:ERAD:18,
title = {Explorando o Paralelismo de Stream em CPU e de Dados em GPU na Aplicação de Filtro Sobel},
author = {Charles M. Stein and Dalvan Griebler},
url = {http://larcc.setrem.com.br/wp-content/uploads/2018/04/LARCC_ERAD_IC_Stein_2018.pdf},
year = {2018},
date = {2018-04-01},
booktitle = {18th Escola Regional de Alto Desempenho do Estado do Rio Grande do Sul (ERAD/RS)},
pages = {137-140},
publisher = {Sociedade Brasileira de Computação},
address = {Porto Alegre, RS, Brazil},
abstract = {O objetivo deste estudo é a paralelização combinada do stream em CPU e dos dados em GPU usando uma aplicação de filtro sobel. Foi realizada uma avaliação do desempenho de OpenCL, OpenACC e CUDA com o algorí-timo de multiplicação de matrizes para escolha da ferramenta a ser usada com a SPar. Concluiu-se que apesar da GPU apresentar um speedup de 11.81x com CUDA, o uso exclusivo da CPU com a SPar é mais vantajoso nesta aplicação.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
O objetivo deste estudo é a paralelização combinada do stream em CPU e dos dados em GPU usando uma aplicação de filtro sobel. Foi realizada uma avaliação do desempenho de OpenCL, OpenACC e CUDA com o algorí-timo de multiplicação de matrizes para escolha da ferramenta a ser usada com a SPar. Concluiu-se que apesar da GPU apresentar um speedup de 11.81x com CUDA, o uso exclusivo da CPU com a SPar é mais vantajoso nesta aplicação. |
| Maliszewski, Anderson M.; Griebler, Dalvan; Schepke, Claudio Desempenho em Instâncias LXC e KVM de Nuvem Privada usando Aplicações Científicas Inproceedings In: 18th Escola Regional de Alto Desempenho do Estado do Rio Grande do Sul (ERAD/RS), pp. 129-132, Sociedade Brasileira de Computação, Porto Alegre, RS, Brazil, 2018. @inproceedings{larcc:cloudtack_lxc_kvm:ERAD:18,
title = {Desempenho em Instâncias LXC e KVM de Nuvem Privada usando Aplicações Científicas},
author = {Anderson M. Maliszewski and Dalvan Griebler and Claudio Schepke},
url = {http://larcc.setrem.com.br/wp-content/uploads/2018/04/LARCC_ERAD_IC_MAliszweski_2018.pdf},
year = {2018},
date = {2018-04-01},
booktitle = {18th Escola Regional de Alto Desempenho do Estado do Rio Grande do Sul (ERAD/RS)},
pages = {129-132},
publisher = {Sociedade Brasileira de Computação},
address = {Porto Alegre, RS, Brazil},
abstract = {As nuvens privadas IaaS oferecem um ambiente atraente para aplicações científicas. Como este ambiente possui camadas adicionais de abstração, alcançar um bom desempenho é um desafio. O objetivo é realizar uma avaliação de desempenho das tecnologias de virtualização baseadas em KVM e LXC gerenciadas pelo CloudStack, usando benchmarks da suite NPB-OMP. Os resultados revelaram que LXC supera KVM em 93,75% dos experimentos.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
As nuvens privadas IaaS oferecem um ambiente atraente para aplicações científicas. Como este ambiente possui camadas adicionais de abstração, alcançar um bom desempenho é um desafio. O objetivo é realizar uma avaliação de desempenho das tecnologias de virtualização baseadas em KVM e LXC gerenciadas pelo CloudStack, usando benchmarks da suite NPB-OMP. Os resultados revelaram que LXC supera KVM em 93,75% dos experimentos. |
| Hoffmann, Renato B.; Griebler, Dalvan; Fernandes, Luiz G. Paralelização de uma Aplicação de Detecção e Eliminação de Ruídos em Streaming de Vídeo com a DSL SPar Inproceedings In: Escola Regional de Alto Desempenho (ERAD-RS), pp. 2, Sociedade Brasileira de Computação (SBC), Porto Alegre, BR, 2018. @inproceedings{HOFFMANN:ERAD:18,
title = {Paralelização de uma Aplicação de Detecção e Eliminação de Ruídos em Streaming de Vídeo com a DSL SPar},
author = {Renato B. Hoffmann and Dalvan Griebler and Luiz G. Fernandes},
url = {https://gmap.pucrs.br/dalvan/papers/2018/CR_ERAD_IC_Hoffmann_2018.pdf},
year = {2018},
date = {2018-04-01},
booktitle = {Escola Regional de Alto Desempenho (ERAD-RS)},
pages = {2},
publisher = {Sociedade Brasileira de Computação (SBC)},
address = {Porto Alegre, BR},
abstract = {Restauração de imagem é uma importante etapa de qualquer sistema de computação gráfica. Este trabalho tem como objetivo apresentar e avaliar o paralelismo de Denoiser, uma aplicação para detecção e eliminação de ruído em streaming de vídeo. Foram avaliados o speed-up e programabilidade das interfaces SPar, Thread Building Blocks e FastFlow. Os resultados mostram que a SPar obteve bons resultados de programabilidade e desempenho.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Restauração de imagem é uma importante etapa de qualquer sistema de computação gráfica. Este trabalho tem como objetivo apresentar e avaliar o paralelismo de Denoiser, uma aplicação para detecção e eliminação de ruído em streaming de vídeo. Foram avaliados o speed-up e programabilidade das interfaces SPar, Thread Building Blocks e FastFlow. Os resultados mostram que a SPar obteve bons resultados de programabilidade e desempenho. |
| Loff, Junior; Griebler, Dalvan; Sandes, Edans; Melo, Alba; Fernandes, Luiz G. Suporte ao Paralelismo Multi-Core com FastFlow e TBB em uma Aplicação de Alinhamento de Sequências de DNA Inproceedings In: Escola Regional de Alto Desempenho (ERAD-RS), pp. 2, Sociedade Brasileira de Computação (SBC), Porto Alegre, BR, 2018. @inproceedings{LOFF:ERAD:18,
title = {Suporte ao Paralelismo Multi-Core com FastFlow e TBB em uma Aplicação de Alinhamento de Sequências de DNA},
author = {Junior Loff and Dalvan Griebler and Edans Sandes and Alba Melo and Luiz G. Fernandes},
url = {https://gmap.pucrs.br/dalvan/papers/2018/CR_ERAD_IC_Loff_2018.pdf},
year = {2018},
date = {2018-04-01},
booktitle = {Escola Regional de Alto Desempenho (ERAD-RS)},
pages = {2},
publisher = {Sociedade Brasileira de Computação (SBC)},
address = {Porto Alegre, BR},
abstract = {Quando uma sequência biológica é obtida, é comum alinhá-la com outra já estudada para determinar suas características. O desafio é processar este alinhamento em tempo útil. Neste trabalho exploramos o paralelismo em uma aplicação de alinhamento de sequências de DNA utilizando as bibliotecas FastFlow e Intel TBB. Os experimentos mostram que a versão TBB obteve até 4% melhor tempo de execução em comparação à versão original em OpenMP.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Quando uma sequência biológica é obtida, é comum alinhá-la com outra já estudada para determinar suas características. O desafio é processar este alinhamento em tempo útil. Neste trabalho exploramos o paralelismo em uma aplicação de alinhamento de sequências de DNA utilizando as bibliotecas FastFlow e Intel TBB. Os experimentos mostram que a versão TBB obteve até 4% melhor tempo de execução em comparação à versão original em OpenMP. |
| Griebler, Dalvan; Loff, Junior; Mencagli, Gabriele; Danelutto, Marco; Fernandes, Luiz Gustavo Efficient NAS Benchmark Kernels with C++ Parallel Programming Inproceedings doi In: 26th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP), pp. 733-740, IEEE, Cambridge, UK, 2018. @inproceedings{GRIEBLER:NAS-CPP:PDP:18,
title = {Efficient NAS Benchmark Kernels with C++ Parallel Programming},
author = {Dalvan Griebler and Junior Loff and Gabriele Mencagli and Marco Danelutto and Luiz Gustavo Fernandes},
url = {https://doi.org/10.1109/PDP2018.2018.00120},
doi = {10.1109/PDP2018.2018.00120},
year = {2018},
date = {2018-03-01},
booktitle = {26th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP)},
pages = {733-740},
publisher = {IEEE},
address = {Cambridge, UK},
series = {PDP'18},
abstract = {Benchmarking is a way to study the performance of new architectures and parallel programming frameworks. Well-established benchmark suites such as the NAS Parallel Benchmarks (NPB) comprise legacy codes that still lack portability to C++ language. As a consequence, a set of high-level and easy-to-use C++ parallel programming frameworks cannot be tested in NPB. Our goal is to describe a C++ porting of the NPB kernels and to analyze the performance achieved by different parallel implementations written using the Intel TBB, OpenMP and FastFlow frameworks for Multi-Cores. The experiments show an efficient code porting from Fortran to C++ and an efficient parallelization on average.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Benchmarking is a way to study the performance of new architectures and parallel programming frameworks. Well-established benchmark suites such as the NAS Parallel Benchmarks (NPB) comprise legacy codes that still lack portability to C++ language. As a consequence, a set of high-level and easy-to-use C++ parallel programming frameworks cannot be tested in NPB. Our goal is to describe a C++ porting of the NPB kernels and to analyze the performance achieved by different parallel implementations written using the Intel TBB, OpenMP and FastFlow frameworks for Multi-Cores. The experiments show an efficient code porting from Fortran to C++ and an efficient parallelization on average. |
 | Griebler, Dalvan; Hoffmann, Renato B.; Danelutto, Marco; Fernandes, Luiz Gustavo High-Level and Productive Stream Parallelism for Dedup, Ferret, and Bzip2 Journal Article doi In: International Journal of Parallel Programming, vol. 47, no. 1, pp. 253-271, 2018, ISSN: 1573-7640. @article{GRIEBLER:IJPP:18,
title = {High-Level and Productive Stream Parallelism for Dedup, Ferret, and Bzip2},
author = {Dalvan Griebler and Renato B. Hoffmann and Marco Danelutto and Luiz Gustavo Fernandes},
url = {https://doi.org/10.1007/s10766-018-0558-x},
doi = {10.1007/s10766-018-0558-x},
issn = {1573-7640},
year = {2018},
date = {2018-02-01},
journal = {International Journal of Parallel Programming},
volume = {47},
number = {1},
pages = {253-271},
publisher = {Springer},
abstract = {Parallel programming has been a challenging task for application programmers. Stream processing is an application domain present in several scientific, enterprise, and financial areas that lack suitable abstractions to exploit parallelism. Our goal is to assess the feasibility of state-of-the-art frameworks/libraries (Pthreads, TBB, and FastFlow) and the SPar domain-specific language for real-world streaming applications (Dedup, Ferret, and Bzip2) targeting multi-core architectures. SPar was specially designed to provide high-level and productive stream parallelism abstractions, supporting programmers with standard C++-11 annotations. For the experiments, we implemented three streaming applications. We discussed SPar’s programmability advantages compared to the frameworks in terms of productivity and structured parallel programming. The results demonstrate that SPar improves productivity and provides the necessary features to achieve similar performances compared to the state-of-the-art.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Parallel programming has been a challenging task for application programmers. Stream processing is an application domain present in several scientific, enterprise, and financial areas that lack suitable abstractions to exploit parallelism. Our goal is to assess the feasibility of state-of-the-art frameworks/libraries (Pthreads, TBB, and FastFlow) and the SPar domain-specific language for real-world streaming applications (Dedup, Ferret, and Bzip2) targeting multi-core architectures. SPar was specially designed to provide high-level and productive stream parallelism abstractions, supporting programmers with standard C++-11 annotations. For the experiments, we implemented three streaming applications. We discussed SPar’s programmability advantages compared to the frameworks in terms of productivity and structured parallel programming. The results demonstrate that SPar improves productivity and provides the necessary features to achieve similar performances compared to the state-of-the-art. |
2017
|
| Griebler, Dalvan; Hoffmann, Renato B.; Loff, Junior; Danelutto, Marco; Fernandes, Luiz G. High-Level and Efficient Stream Parallelism on Multi-core Systems with SPar for Data Compression Applications Inproceedings In: XVIII Simpósio em Sistemas Computacionais de Alto Desempenho, pp. 16-27, SBC, Campinas, SP, Brasil, 2017. @inproceedings{GRIEBLER:WSCAD:17,
title = {High-Level and Efficient Stream Parallelism on Multi-core Systems with SPar for Data Compression Applications},
author = {Dalvan Griebler and Renato B. Hoffmann and Junior Loff and Marco Danelutto and Luiz G. Fernandes},
url = {https://gmap.pucrs.br/dalvan/papers/2017/CR_WSCAD_2017.pdf},
year = {2017},
date = {2017-10-01},
booktitle = {XVIII Simpósio em Sistemas Computacionais de Alto Desempenho},
pages = {16-27},
publisher = {SBC},
address = {Campinas, SP, Brasil},
abstract = {The stream processing domain is present in several real-world applications that are running on multi-core systems. In this paper, we focus on data compression applications that are an important sub-set of this domain. Our main goal is to assess the programmability and efficiency of domain-specific language called SPar. It was specially designed for expressing stream parallelism and it promises higher-level parallelism abstractions without significant performance losses. Therefore, we parallelized Lzip and Bzip2 compressors with SPar and compared with state-of-the-art frameworks. The results revealed that SPar is able to efficiently exploit stream parallelism as well as provide suitable abstractions with less code intrusion and code re-factoring.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The stream processing domain is present in several real-world applications that are running on multi-core systems. In this paper, we focus on data compression applications that are an important sub-set of this domain. Our main goal is to assess the programmability and efficiency of domain-specific language called SPar. It was specially designed for expressing stream parallelism and it promises higher-level parallelism abstractions without significant performance losses. Therefore, we parallelized Lzip and Bzip2 compressors with SPar and compared with state-of-the-art frameworks. The results revealed that SPar is able to efficiently exploit stream parallelism as well as provide suitable abstractions with less code intrusion and code re-factoring. |
| Griebler, Dalvan; Fernandes, Luiz Gustavo Towards Distributed Parallel Programming Support for the SPar DSL Inproceedings doi In: Parallel Computing is Everywhere, Proceedings of the International
Conference on Parallel Computing, pp. 563-572, IOS Press, Bologna, Italy, 2017. @inproceedings{GRIEBLER:PARCO:17,
title = {Towards Distributed Parallel Programming Support for the SPar DSL},
author = {Dalvan Griebler and Luiz Gustavo Fernandes},
url = {https://doi.org/10.3233/978-1-61499-843-3-563},
doi = {10.3233/978-1-61499-843-3-563},
year = {2017},
date = {2017-09-01},
booktitle = {Parallel Computing is Everywhere, Proceedings of the International
Conference on Parallel Computing},
pages = {563-572},
publisher = {IOS Press},
address = {Bologna, Italy},
series = {ParCo'17},
abstract = {SPar was originally designed to provide high-level abstractions for stream parallelism in C++ programs targeting multi-core systems. This work proposes distributed parallel programming support for SPar targeting cluster environments. The goal is to preserve the original semantics while source-to-source code transformations will be turned into MPI (Message Passing Interface) parallel code. The results of the experiments presented in the paper demonstrate improved programmability without significant performance losses.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
SPar was originally designed to provide high-level abstractions for stream parallelism in C++ programs targeting multi-core systems. This work proposes distributed parallel programming support for SPar targeting cluster environments. The goal is to preserve the original semantics while source-to-source code transformations will be turned into MPI (Message Passing Interface) parallel code. The results of the experiments presented in the paper demonstrate improved programmability without significant performance losses. |
| Griebler, Dalvan; Hoffmann, Renato B.; Danelutto, Marco; Fernandes, Luiz Gustavo Higher-Level Parallelism Abstractions for Video Applications with SPar Inproceedings doi In: Parallel Computing is Everywhere, Proceedings of the International Conference on Parallel Computing, pp. 698-707, IOS Press, Bologna, Italy, 2017. @inproceedings{GRIEBLER:REPARA:17,
title = {Higher-Level Parallelism Abstractions for Video Applications with SPar},
author = {Dalvan Griebler and Renato B. Hoffmann and Marco Danelutto and Luiz Gustavo Fernandes},
url = {https://doi.org/10.3233/978-1-61499-843-3-698},
doi = {10.3233/978-1-61499-843-3-698},
year = {2017},
date = {2017-09-01},
booktitle = {Parallel Computing is Everywhere, Proceedings of the International Conference on Parallel Computing},
pages = {698-707},
publisher = {IOS Press},
address = {Bologna, Italy},
series = {ParCo'17},
abstract = {SPar is a Domain-Specific Language (DSL) designed to provide high-level parallel programming abstractions for streaming applications. Video processing application domain requires parallel processing to extract and analyze information quickly. When using state-of-the-art frameworks such as FastFlow and TBB, the application programmer has to manage source code re-factoring and performance optimization to implement parallelism efficiently. Our goal is to make this process easier for programmers through SPar. Thus we assess SPar's programming language and its performance in traditional video applications. We also discuss different implementations compared to the ones of SPar. Results demonstrate that SPar maintains the sequential code structure, is less code intrusive, and provides higher-level programming abstractions without introducing notable performance losses. Therefore, it represents a good choice for application programmers from the video processing domain.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
SPar is a Domain-Specific Language (DSL) designed to provide high-level parallel programming abstractions for streaming applications. Video processing application domain requires parallel processing to extract and analyze information quickly. When using state-of-the-art frameworks such as FastFlow and TBB, the application programmer has to manage source code re-factoring and performance optimization to implement parallelism efficiently. Our goal is to make this process easier for programmers through SPar. Thus we assess SPar's programming language and its performance in traditional video applications. We also discuss different implementations compared to the ones of SPar. Results demonstrate that SPar maintains the sequential code structure, is less code intrusive, and provides higher-level programming abstractions without introducing notable performance losses. Therefore, it represents a good choice for application programmers from the video processing domain. |
| Leiria, Raul; Vogel, Adriano; Griebler, Dalvan; Schepke, Claudio Uma Proposta para o Monitoramento Energético de Nuvens Computacionais Privadas no Zabbix Inproceedings In: 15th Escola Regional de Redes de Computadores (ERRC), pp. 1-4, Sociedade Brasileira de Computação, Santa Maria, BR, 2017. @inproceedings{larcc:zabbix_energy_cloud:ERRC:17,
title = {Uma Proposta para o Monitoramento Energético de Nuvens Computacionais Privadas no Zabbix},
author = {Raul Leiria and Adriano Vogel and Dalvan Griebler and Claudio Schepke},
url = {http://larcc.setrem.com.br/wp-content/uploads/2018/02/CR_ERRC_Leiria_2017.pdf},
year = {2017},
date = {2017-09-01},
booktitle = {15th Escola Regional de Redes de Computadores (ERRC)},
pages = {1-4},
publisher = {Sociedade Brasileira de Computação},
address = {Santa Maria, BR},
abstract = {In the last years, cloud computing has been consolidated as a new computational paradigm due to its widespread adoption. Proportionally, this leverages to the increasing of the power consumption by data centers. As a consequence, there is a witnessed about the growing demand for monitoring the power consumption on computational infrastructures. Therefore, in this work is proposed a mechanism to monitor the cloud computing power draw through the Zabbix open-source networking monitoring tool.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
In the last years, cloud computing has been consolidated as a new computational paradigm due to its widespread adoption. Proportionally, this leverages to the increasing of the power consumption by data centers. As a consequence, there is a witnessed about the growing demand for monitoring the power consumption on computational infrastructures. Therefore, in this work is proposed a mechanism to monitor the cloud computing power draw through the Zabbix open-source networking monitoring tool. |
| Ledur, Cleverson; Griebler, Dalvan; Manssour, Isabel; Fernandes, Luiz Gustavo A High-Level DSL for Geospatial Visualizations with Multi-core Parallelism Support Inproceedings doi In: 41th IEEE Computer Society Signature Conference on Computers, Software and Applications, pp. 298-304, IEEE, Torino, Italy, 2017. @inproceedings{LEDUR:COMPSAC:17,
title = {A High-Level DSL for Geospatial Visualizations with Multi-core Parallelism Support},
author = {Cleverson Ledur and Dalvan Griebler and Isabel Manssour and Luiz Gustavo Fernandes},
url = {https://doi.org/10.1109/COMPSAC.2017.18},
doi = {10.1109/COMPSAC.2017.18},
year = {2017},
date = {2017-07-01},
booktitle = {41th IEEE Computer Society Signature Conference on Computers, Software and Applications},
pages = {298-304},
publisher = {IEEE},
address = {Torino, Italy},
series = {COMPSAC'17},
abstract = {The amount of data generated worldwide associated with geolocalization has exponentially increased over the last decade due to social networks, population demographics, and the popularization of Global Positioning Systems. Several methods for geovisualization have already been developed, but many of them are focused on a specific application or require learning a variety of tools and programming languages. It becomes even more difficult when users have to manage a large amount of data because state-of-the-art alternatives require the use of third-party pre-processing tools. We present a novel Domain-Specific Language (DSL), which focuses on large data geovisualizations. Through a compiler, we support automatic visualization generations and data pre-processing. The system takes advantage of multi-core parallelism to speed-up data pre-processing abstractly. Our experiments were designated to highlight the programming effort and performance of our DSL. The results have shown a considerable programming effort reduction and efficient parallelism support with respect to the sequential version.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The amount of data generated worldwide associated with geolocalization has exponentially increased over the last decade due to social networks, population demographics, and the popularization of Global Positioning Systems. Several methods for geovisualization have already been developed, but many of them are focused on a specific application or require learning a variety of tools and programming languages. It becomes even more difficult when users have to manage a large amount of data because state-of-the-art alternatives require the use of third-party pre-processing tools. We present a novel Domain-Specific Language (DSL), which focuses on large data geovisualizations. Through a compiler, we support automatic visualization generations and data pre-processing. The system takes advantage of multi-core parallelism to speed-up data pre-processing abstractly. Our experiments were designated to highlight the programming effort and performance of our DSL. The results have shown a considerable programming effort reduction and efficient parallelism support with respect to the sequential version. |
| Rista, Cassiano; Griebler, Dalvan; Maron, Carlos A. F.; Fernandes, Luiz Gustavo Improving the Network Performance of a Container-Based Cloud Environment for Hadoop Systems Inproceedings doi In: International Conference on High Performance Computing & Simulation (HPCS), pp. 619-626, IEEE, Genoa, Italy, 2017. @inproceedings{larcc:link_aggregation:HPCS:2017,
title = {Improving the Network Performance of a Container-Based Cloud Environment for Hadoop Systems},
author = {Cassiano Rista and Dalvan Griebler and Carlos A. F. Maron and Luiz Gustavo Fernandes},
url = {http://ieeexplore.ieee.org/document/8035136/},
doi = {10.1109/HPCS.2017.97},
year = {2017},
date = {2017-07-01},
booktitle = {International Conference on High Performance Computing & Simulation (HPCS)},
pages = {619-626},
publisher = {IEEE},
address = {Genoa, Italy},
series = {HPCS'17},
abstract = {Cloud computing has emerged as an important paradigm to improve resource utilization, efficiency, flexibility, and the pay-per-use billing structure. However, cloud platforms cause performance degradations due to their virtualization layer and may not be appropriate for the requirements of high-performance applications, such as big data. This paper tackles the problem of improving network performance in container-based cloud instances to create a viable alternative to run network intensive Hadoop applications. Our approach consists of deploying link aggregation via the IEEE 802.3ad standard to increase the available bandwidth and using LXC (Linux Container) cloud instances to create a Hadoop cluster. In order to evaluate the efficiency of our approach and the overhead added by the container-based cloud environment, we ran a set of experiments to measure throughput, latency, bandwidth utilization, and completion times. The results prove that our approach adds minimal overhead in cloud environment as well as increases throughput and reduces latency. Moreover, our approach demonstrates a suitable alternative for running Hadoop applications, reducing completion times up to 33.73%},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Cloud computing has emerged as an important paradigm to improve resource utilization, efficiency, flexibility, and the pay-per-use billing structure. However, cloud platforms cause performance degradations due to their virtualization layer and may not be appropriate for the requirements of high-performance applications, such as big data. This paper tackles the problem of improving network performance in container-based cloud instances to create a viable alternative to run network intensive Hadoop applications. Our approach consists of deploying link aggregation via the IEEE 802.3ad standard to increase the available bandwidth and using LXC (Linux Container) cloud instances to create a Hadoop cluster. In order to evaluate the efficiency of our approach and the overhead added by the container-based cloud environment, we ran a set of experiments to measure throughput, latency, bandwidth utilization, and completion times. The results prove that our approach adds minimal overhead in cloud environment as well as increases throughput and reduces latency. Moreover, our approach demonstrates a suitable alternative for running Hadoop applications, reducing completion times up to 33.73% |
| Rockenbach, Dinei A.; Anderle, Nadine; Griebler, Dalvan; Souza, Samuel Estudo Comparativo de Banco de Dados Chave-Valor com Armazenamento em Memória Inproceedings In: 13th Escola Regional de Banco de Dados (ERBD), pp. 1-4, Sociedade Brasileira de Computação, Passo Fundo, BR, 2017. @inproceedings{larcc:database_keyvalue:ERBD:17,
title = {Estudo Comparativo de Banco de Dados Chave-Valor com Armazenamento em Memória},
author = {Dinei A. Rockenbach and Nadine Anderle and Dalvan Griebler and Samuel Souza},
url = {http://larcc.setrem.com.br/wp-content/uploads/2017/03/ANDERLE_ERBD_2017.pdf},
year = {2017},
date = {2017-04-01},
booktitle = {13th Escola Regional de Banco de Dados (ERBD)},
pages = {1-4},
publisher = {Sociedade Brasileira de Computação},
address = {Passo Fundo, BR},
abstract = {Key-value databases emerge to address relational databases' limitations and with the increasing capacity of RAM memory it is possible to offer greater performance and versatility in data storage and processing. The objective is to perform a comparative study of key-value databases with memory storage Redis, Memcached, Voldemort, Aerospike, Hazelcast and Riak KV. Thus, the work contributed to an analysis of different databases and with results that qualitatively demonstrated the characteristics and pointed out the main advantages.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Key-value databases emerge to address relational databases' limitations and with the increasing capacity of RAM memory it is possible to offer greater performance and versatility in data storage and processing. The objective is to perform a comparative study of key-value databases with memory storage Redis, Memcached, Voldemort, Aerospike, Hazelcast and Riak KV. Thus, the work contributed to an analysis of different databases and with results that qualitatively demonstrated the characteristics and pointed out the main advantages. |