2022 |
Vogel, Adriano Self-adaptive abstractions for efficient high-level parallel computing in multi-cores PhD Thesis School of Technology - PUCRS, 2022. @phdthesis{VOGEL:PHD_PUCRS:22, title = {Self-adaptive abstractions for efficient high-level parallel computing in multi-cores}, author = {Adriano Vogel}, url = {https://tede2.pucrs.br/tede2/handle/tede/10232}, year = {2022}, date = {2022-05-01}, address = {Porto Alegre, Brazil}, school = {School of Technology - PUCRS}, abstract = {Nowadays, a significant part of computing systems and real-world applications demand parallelism to accelerate their executions. Although high-level and structured parallel programming aims to facilitate parallelism exploitation, there are still issues to be addressed to improve existing parallel programming abstractions. Usually, application developers still have to set non-intuitive or complex parallelism configurations. In this context, self-adaptation is a potential alternative to provide a higher-level of autonomic abstractions and runtime responsiveness in parallel executions. However, a recurrent problem is that self-adaptation is still limited in terms of flexibility, efficiency, and abstractions. For instance, there is a lack of mechanisms to apply adaptation actions and efficient decision-making strategies to decide which configurations to be enforced at run-time. In this work, we are interested in abstractions achievable with self-adaptation transparently managing the executions while the parallel programs are running (at run-time). Our main goals are to increase the adaptation space to be more representative of real-world applications and make self-adaptation more efficient with comprehensive evaluation methodologies, which can provide use-cases demonstrating the true potentials of self-adaptation. Therefore, this doctoral dissertation provides the following scientific contributions: I) An Systematic Literature Review (SLR) providing a taxonomy of the state-of-the-art. II) A conceptual framework to support designing and abstracting the decision-making process within self-adaptive solutions, such a conceptual framework is then employed in the technical contributions to assist in making the solutions more modular and potentially generalizable. III) Mechanisms and strategies for self-adaptive replicas in applications with single and multiple parallel stages, supporting multiple customizable non-functional requirements. IV) Mechanism, strategy, and optimizations for self-adaptation of Parallel Patterns/applications’ graphs topologies. We apply the proposed solutions to the context of stream processing applications, a representative paradigm present in several real-world applications that compute data flowing in the form of streams (e.g., video feeds, image, and data analytics). A part of the proposed solutions is evaluated with SPar and another part with the FastFlow programming framework. The results demonstrate that self-adaptation can provide efficient parallelism abstractions and autonomous responsiveness at run-time, yet achieve a competitive performance w.r.t. the best static executions. Moreover, when appropriate, we compare state-of-the-art solutions and demonstrate that our highly optimized decision-making strategies achieve significant performance and efficiency gains.}, keywords = {}, pubstate = {published}, tppubtype = {phdthesis} } Nowadays, a significant part of computing systems and real-world applications demand parallelism to accelerate their executions. Although high-level and structured parallel programming aims to facilitate parallelism exploitation, there are still issues to be addressed to improve existing parallel programming abstractions. Usually, application developers still have to set non-intuitive or complex parallelism configurations. In this context, self-adaptation is a potential alternative to provide a higher-level of autonomic abstractions and runtime responsiveness in parallel executions. However, a recurrent problem is that self-adaptation is still limited in terms of flexibility, efficiency, and abstractions. For instance, there is a lack of mechanisms to apply adaptation actions and efficient decision-making strategies to decide which configurations to be enforced at run-time. In this work, we are interested in abstractions achievable with self-adaptation transparently managing the executions while the parallel programs are running (at run-time). Our main goals are to increase the adaptation space to be more representative of real-world applications and make self-adaptation more efficient with comprehensive evaluation methodologies, which can provide use-cases demonstrating the true potentials of self-adaptation. Therefore, this doctoral dissertation provides the following scientific contributions: I) An Systematic Literature Review (SLR) providing a taxonomy of the state-of-the-art. II) A conceptual framework to support designing and abstracting the decision-making process within self-adaptive solutions, such a conceptual framework is then employed in the technical contributions to assist in making the solutions more modular and potentially generalizable. III) Mechanisms and strategies for self-adaptive replicas in applications with single and multiple parallel stages, supporting multiple customizable non-functional requirements. IV) Mechanism, strategy, and optimizations for self-adaptation of Parallel Patterns/applications’ graphs topologies. We apply the proposed solutions to the context of stream processing applications, a representative paradigm present in several real-world applications that compute data flowing in the form of streams (e.g., video feeds, image, and data analytics). A part of the proposed solutions is evaluated with SPar and another part with the FastFlow programming framework. The results demonstrate that self-adaptation can provide efficient parallelism abstractions and autonomous responsiveness at run-time, yet achieve a competitive performance w.r.t. the best static executions. Moreover, when appropriate, we compare state-of-the-art solutions and demonstrate that our highly optimized decision-making strategies achieve significant performance and efficiency gains. |
Vogel, Adriano Self-adaptive abstractions for efficient high-level parallel computing in multi-cores PhD Thesis Computer Science Department - University of Pisa, 2022. @phdthesis{VOGEL:PHD_PISA:22, title = {Self-adaptive abstractions for efficient high-level parallel computing in multi-cores}, author = {Adriano Vogel}, url = {https://etd.adm.unipi.it/theses/available/etd-04142022-142258/unrestricted/Vogel_PhD_Dissertation_UNIPI.pdf}, year = {2022}, date = {2022-05-01}, address = {Pisa, Italy}, school = {Computer Science Department - University of Pisa}, abstract = {Nowadays, a significant part of computing systems and real-world applications demand parallelism to accelerate their executions. Although high-level and structured parallel programming aims to facilitate parallelism exploitation, there are still issues to be addressed to improve existing parallel programming abstractions. Usually, application developers still have to set non-intuitive or complex parallelism configurations. In this context, self-adaptation is a potential alternative to provide a higher-level of autonomic abstractions and runtime responsiveness in parallel executions. However, a recurrent problem is that self-adaptation is still limited in terms of flexibility, efficiency, and abstractions. For instance, there is a lack of mechanisms to apply adaptation actions and efficient decision-making strategies to decide which configurations to be enforced at run-time. In this work, we are interested in abstractions achievable with self-adaptation transparently managing the executions while the parallel programs are running (at run-time). Our main goals are to increase the adaptation space to be more representative of real-world applications and make self-adaptation more efficient with comprehensive evaluation methodologies, which can provide use-cases demonstrating the true potentials of self-adaptation. Therefore, this doctoral dissertation provides the following scientific contributions: I) An Systematic Literature Review (SLR) providing a taxonomy of the state-of-the-art. II) A conceptual framework to support designing and abstracting the decision-making process within self-adaptive solutions, such a conceptual framework is then employed in the technical contributions to assist in making the solutions more modular and potentially generalizable. III) Mechanisms and strategies for self-adaptive replicas in applications with single and multiple parallel stages, supporting multiple customizable non-functional requirements. IV) Mechanism, strategy, and optimizations for self-adaptation of Parallel Patterns/applications’ graphs topologies. We apply the proposed solutions to the context of stream processing applications, a representative paradigm present in several real-world applications that compute data flowing in the form of streams (e.g., video feeds, image, and data analytics). A part of the proposed solutions is evaluated with SPar and another part with the FastFlow programming framework. The results demonstrate that self-adaptation can provide efficient parallelism abstractions and autonomous responsiveness at run-time, yet achieve a competitive performance w.r.t. the best static executions. Moreover, when appropriate, we compare state-of-the-art solutions and demonstrate that our highly optimized decision-making strategies achieve significant performance and efficiency gains.}, keywords = {}, pubstate = {published}, tppubtype = {phdthesis} } Nowadays, a significant part of computing systems and real-world applications demand parallelism to accelerate their executions. Although high-level and structured parallel programming aims to facilitate parallelism exploitation, there are still issues to be addressed to improve existing parallel programming abstractions. Usually, application developers still have to set non-intuitive or complex parallelism configurations. In this context, self-adaptation is a potential alternative to provide a higher-level of autonomic abstractions and runtime responsiveness in parallel executions. However, a recurrent problem is that self-adaptation is still limited in terms of flexibility, efficiency, and abstractions. For instance, there is a lack of mechanisms to apply adaptation actions and efficient decision-making strategies to decide which configurations to be enforced at run-time. In this work, we are interested in abstractions achievable with self-adaptation transparently managing the executions while the parallel programs are running (at run-time). Our main goals are to increase the adaptation space to be more representative of real-world applications and make self-adaptation more efficient with comprehensive evaluation methodologies, which can provide use-cases demonstrating the true potentials of self-adaptation. Therefore, this doctoral dissertation provides the following scientific contributions: I) An Systematic Literature Review (SLR) providing a taxonomy of the state-of-the-art. II) A conceptual framework to support designing and abstracting the decision-making process within self-adaptive solutions, such a conceptual framework is then employed in the technical contributions to assist in making the solutions more modular and potentially generalizable. III) Mechanisms and strategies for self-adaptive replicas in applications with single and multiple parallel stages, supporting multiple customizable non-functional requirements. IV) Mechanism, strategy, and optimizations for self-adaptation of Parallel Patterns/applications’ graphs topologies. We apply the proposed solutions to the context of stream processing applications, a representative paradigm present in several real-world applications that compute data flowing in the form of streams (e.g., video feeds, image, and data analytics). A part of the proposed solutions is evaluated with SPar and another part with the FastFlow programming framework. The results demonstrate that self-adaptation can provide efficient parallelism abstractions and autonomous responsiveness at run-time, yet achieve a competitive performance w.r.t. the best static executions. Moreover, when appropriate, we compare state-of-the-art solutions and demonstrate that our highly optimized decision-making strategies achieve significant performance and efficiency gains. |
Andrade, Gabriella; Griebler, Dalvan; Fernandes, Luiz Gustavo Avaliação do Esforço de Programação em GPU: Estudo Piloto Inproceedings doi Anais da XXII Escola Regional de Alto Desempenho da Região Sul, pp. 95–96, Sociedade Brasileira de Computação, Curitiba, Brazil, 2022. @inproceedings{ANDRADE:ERAD:22, title = {Avaliação do Esforço de Programação em GPU: Estudo Piloto}, author = {Gabriella Andrade and Dalvan Griebler and Luiz Gustavo Fernandes}, url = {https://doi.org/10.5753/eradrs.2022.19179}, doi = {10.5753/eradrs.2022.19179}, year = {2022}, date = {2022-04-01}, booktitle = {Anais da XXII Escola Regional de Alto Desempenho da Região Sul}, pages = {95--96}, publisher = {Sociedade Brasileira de Computação}, address = {Curitiba, Brazil}, abstract = {O desenvolvimento de aplicações para GPU não é uma tarefa fácil, pois exige um maior conhecimento da arquitetura. Neste trabalho realizamos um estudo piloto para avaliar o esforço de programadores não-especialistas ao desenvolver aplicações para GPU. Os resultados revelaram que a GSParLib requer menos esforço em relação as demais interfaces de programação paralela. Entretanto, mais investigações são necessárias a fim de complementar o estudo.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } O desenvolvimento de aplicações para GPU não é uma tarefa fácil, pois exige um maior conhecimento da arquitetura. Neste trabalho realizamos um estudo piloto para avaliar o esforço de programadores não-especialistas ao desenvolver aplicações para GPU. Os resultados revelaram que a GSParLib requer menos esforço em relação as demais interfaces de programação paralela. Entretanto, mais investigações são necessárias a fim de complementar o estudo. |
Garcia, Adriano Marques; Griebler, Dalvan; Schepke, Claudio; Fernandes, Luiz Gustavo Um Framework para Criar Benchmarks de Aplicações Paralelas de Stream Inproceedings doi Anais da XXII Escola Regional de Alto Desempenho da Região Sul, pp. 97–98, Sociedade Brasileira de Computação, Curitiba, Brazil, 2022. @inproceedings{GARCIA:ERAD:22, title = {Um Framework para Criar Benchmarks de Aplicações Paralelas de Stream}, author = {Adriano Marques Garcia and Dalvan Griebler and Claudio Schepke and Luiz Gustavo Fernandes}, url = {https://doi.org/10.5753/eradrs.2022.19180}, doi = {10.5753/eradrs.2022.19180}, year = {2022}, date = {2022-04-01}, booktitle = {Anais da XXII Escola Regional de Alto Desempenho da Região Sul}, pages = {97--98}, publisher = {Sociedade Brasileira de Computação}, address = {Curitiba, Brazil}, abstract = {Este trabalho apresenta o SPBench, um framework para o desenvolvimento de benchmarks de processamento de stream em C++. O SPBench fornece um conjunto de aplicações realísticas através de abstrações de alto nível e permite customizações nos dados de entrada e métricas de desempenho.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } Este trabalho apresenta o SPBench, um framework para o desenvolvimento de benchmarks de processamento de stream em C++. O SPBench fornece um conjunto de aplicações realísticas através de abstrações de alto nível e permite customizações nos dados de entrada e métricas de desempenho. |
Garcia, Adriano Marques; Griebler, Dalvan; Schepke, Claudio; Fernandes, Luiz Gustavo Evaluating Micro-batch and Data Frequency for Stream Processing Applications on Multi-cores Inproceedings doi 30th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP), pp. 10-17, IEEE, Valladolid, Spain, 2022. @inproceedings{GARCIA:PDP:22, title = {Evaluating Micro-batch and Data Frequency for Stream Processing Applications on Multi-cores}, author = {Adriano Marques Garcia and Dalvan Griebler and Claudio Schepke and Luiz Gustavo Fernandes}, url = {https://doi.org/10.1109/PDP55904.2022.00011}, doi = {10.1109/PDP55904.2022.00011}, year = {2022}, date = {2022-04-01}, booktitle = {30th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP)}, pages = {10-17}, publisher = {IEEE}, address = {Valladolid, Spain}, series = {PDP'22}, abstract = {In stream processing, data arrives constantly and is often unpredictable. It can show large fluctuations in arrival frequency, size, complexity, and other factors. These fluctuations can strongly impact application latency and throughput, which are critical factors in this domain. Therefore, there is a significant amount of research on self-adaptive techniques involving elasticity or micro-batching as a way to mitigate this impact. However, there is a lack of benchmarks and tools for helping researchers to investigate micro-batching and data stream frequency implications. In this paper, we extend a benchmarking framework to support dynamic micro-batching and data stream frequency management. We used it to create custom benchmarks and compare latency and throughput aspects from two different parallel libraries. We validate our solution through an extensive analysis of the impact of micro-batching and data stream frequency on stream processing applications using Intel TBB and FastFlow, which are two libraries that leverage stream parallelism on multi-core architectures. Our results demonstrated up to 33% throughput gain over latency using micro-batches. Additionally, while TBB ensures lower latency, FastFlow ensures higher throughput in the parallel applications for different data stream frequency configurations.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } In stream processing, data arrives constantly and is often unpredictable. It can show large fluctuations in arrival frequency, size, complexity, and other factors. These fluctuations can strongly impact application latency and throughput, which are critical factors in this domain. Therefore, there is a significant amount of research on self-adaptive techniques involving elasticity or micro-batching as a way to mitigate this impact. However, there is a lack of benchmarks and tools for helping researchers to investigate micro-batching and data stream frequency implications. In this paper, we extend a benchmarking framework to support dynamic micro-batching and data stream frequency management. We used it to create custom benchmarks and compare latency and throughput aspects from two different parallel libraries. We validate our solution through an extensive analysis of the impact of micro-batching and data stream frequency on stream processing applications using Intel TBB and FastFlow, which are two libraries that leverage stream parallelism on multi-core architectures. Our results demonstrated up to 33% throughput gain over latency using micro-batches. Additionally, while TBB ensures lower latency, FastFlow ensures higher throughput in the parallel applications for different data stream frequency configurations. |
Parallel Applications Modelling Group
GMAP is a research group at the Pontifical Catholic University of Rio Grande do Sul (PUCRS). Historically, the group has conducted several types of research on modeling and adapting robust, real-world applications from different domains (physics, mathematics, geology, image processing, biology, aerospace, and many others) to run efficiently on High-Performance Computing (HPC) architectures, such as Clusters.
In the last decade, new abstractions of parallelism are being created through domain-specific languages (DSLs), libraries, and frameworks for the next generation of computer algorithms and architectures, such as embedded hardware and servers with accelerators like Graphics Processing Units (GPUs) or Field-Programmable Gate Array (FPGAs). This has been applied to stream processing and data science-oriented applications. Concomitantly, since 2018, research is being conducted using artificial intelligence to optimize applications in the areas of Medicine, Ecology, Industry, Agriculture, Education, Smart Cities, and others.
Research Lines
Applied Data Science
This line of research aims to apply and develop methods for knowledge extraction from databases, management of large volumes of data, explore artificial intelligence algorithms for problem-solving in Medicine, Ecology, Industry, Smart Cities, Agriculture, and other application domains. Data-intensive processing platforms are also enhanced for specific needs.
Parallelism Abstractions
The research line HSPA (High-level and Structured Parallelism Abstractions) aims to create programming interfaces for the user/programmer who is not able in dealing with the parallel programming paradigm. The idea is to offer a higher level of abstraction, where the performance of applications is not compromised. The interfaces developed in this research line go toward specific domains that can later extend to other areas. The scope of the study is broad as regards the use of technologies for the development of the interface and parallelism.
Parallel Application Modeling
The Parallel Applications Modelling research line is centered on the study of high-performance solutions for problems that originated in other areas of knowledge (Medicine, Biology, Geology, Physics, etc). The development of high-performance solutions for problems requires a grasp of many levels of specialized knowledge. Besides an analysis of the algorithmic complexity of the problem to be parallelized, it is also necessary to have knowledge about the developing environments and program tests, parallel programming techniques, performance evaluations, and high-performance architectures.
Team
Prof. Dr. Luiz Gustavo Leão Fernandes
General Coordinator
Associate Professor at Polytechnic School, at the Graduate Program in Computer Science, and coordinator of the Stricto Sensu Programs of the Pro-Rectory of Research and Graduate Studies at PUCRS. Founder of GMAP. He works in the management of research projects and student orientation.
Prof. Dr. Dalvan Griebler
Research Coordinator
Associate Professor at the Polytechnic School and Computer Science Graduate Program at PUCRS. He also works with the management and development of research projects and student orientation.
Last Papers
2022 |
Self-adaptive abstractions for efficient high-level parallel computing in multi-cores PhD Thesis School of Technology - PUCRS, 2022. |
Self-adaptive abstractions for efficient high-level parallel computing in multi-cores PhD Thesis Computer Science Department - University of Pisa, 2022. |
Avaliação do Esforço de Programação em GPU: Estudo Piloto Inproceedings doi Anais da XXII Escola Regional de Alto Desempenho da Região Sul, pp. 95–96, Sociedade Brasileira de Computação, Curitiba, Brazil, 2022. |
Um Framework para Criar Benchmarks de Aplicações Paralelas de Stream Inproceedings doi Anais da XXII Escola Regional de Alto Desempenho da Região Sul, pp. 97–98, Sociedade Brasileira de Computação, Curitiba, Brazil, 2022. |
Evaluating Micro-batch and Data Frequency for Stream Processing Applications on Multi-cores Inproceedings doi 30th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP), pp. 10-17, IEEE, Valladolid, Spain, 2022. |
Projects
GMAP is involved in various research projects in many areas of computing. To see a list of all completed or ongoing projects, please visit the Projects page.
Software
To see a list of the software developed by the researchers from GMAP, please visit the Software page.
Last News
GMAP members win second place in the ERAD/RS 2022 parallel programming marathon.
During the XXII Regional School of High Performance of the South Region (ERAD/RS), the fifth edition of the Parallel Programming Marathon took place to promote knowledge in parallel and distributed programming. The target audience is undergraduate and graduate students of Computer Science, Computer Engineering, Information Systems, and related courses. This year, undergraduate student Caetano Muller,…
GMAP at ERAD-RS 2022
The Regional High-Performance School (ERAD/RS) takes place annually. Its main objectives are to qualify future professionals from southern Brazil in the areas that compose the high-performance processing, provide a regular forum in which recent advances in these areas can be presented, and discuss ways of teaching high-performance processing in universities. ERAD/RS has presentation tracks for…
Gabriell Araujo was approved in his master thesis
On April 4th, Gabriell Alves de Araujo successfully defended his master’s thesis “Data and Stream Parallelism Optimization on GPUs” at the School of Technology of the Pontifical Catholic University of Rio Grande do Sul (PUCRS). Gabriell is currently a Ph.D. candidate at GMAP. The defense was done through video conference due to the events of…
Contact us!
If you have any questions or proposals, whether you are a student or a company seeking partnership for research projects, we are available!
Or, feel free to use the form below to contact us.