Adaptive loops with kaapi on multicore and grid

Danjean, Vincent; Gillard, R.; Guelton, Serge; Roch, Jean‐Louis; Roche, Thomas

doi:10.1145/1278177.1278185

Cited by 9 publications

(8 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Indeed, the latter depends on data computed during the iterations. To handle this issue we propose an adaptive parallel algorithm [5,24] and use the Kaapi library [6,16]. Its expressiveness in an adaptive setting guided our choice, together with the possibility to work on heterogenous networks.…”

Section: Parallelizationmentioning

confidence: 99%

“…In the following, we denote by T seq the time of the sequential execution and by T p the time of the parallel execution for p = 8 or p = 16 cores. All the matrices are from "Sparse Integer Matrix Collection" (SIMC) 6 . Table 3 gives the performance of the parallel computation of the determinant for small invertible matrices (less than a second) and larger ones (an hour CPU) of the SIMC/SPG and SIMC/Trefethen collections.…”

Section: Efficiencymentioning

confidence: 99%

See 1 more Smart Citation

Generic design of Chinese remaindering schemes

Dumas

Gautier

Roch

2010

Proceedings of the 4th International Workshop on Parallel and Symbolic Computation

Self Cite

View full text Add to dashboard Cite

We propose a generic design for Chinese remainder algorithms. A Chinese remainder computation consists in reconstructing an integer value from its residues modulo non coprime integers. We also propose an efficient linear data structure, a radix ladder, for the intermediate storage and computations. Our design is structured into three main modules: a black box residue computation in charge of computing each residue; a Chinese remaindering controller in charge of launching the computation and of the termination decision; an integer builder in charge of the reconstruction computation. We then show that this design enables many different forms of Chinese remaindering (e.g. deterministic, early terminated, distributed, etc.), easy comparisons between these forms and e.g. user-transparent parallelism at different parallel grains.

show abstract

Section: Parallelizationmentioning

confidence: 99%

Section: Efficiencymentioning

confidence: 99%

Generic design of Chinese remaindering schemes

Dumas

Gautier

Roch

2010

Proceedings of the 4th International Workshop on Parallel and Symbolic Computation

Self Cite

View full text Add to dashboard Cite

show abstract

“…Cipher parallelism On CPU, Roche et al [14] propose to use a block cipher in counter chaining mode, a mode that is well-suited for parallelization in addition to feature a high security level. The authors use a very effective method based on work-stealing and loop rescheduling for DES encryption.…”

Section: Related Workmentioning

confidence: 99%

Parallel arithmetic encryption for high-bandwidth communications on multicore/GPGPU platforms

Jacquin

Roca

Roch

et al. 2010

Proceedings of the 4th International Workshop on Parallel and Symbolic Computation

Self Cite

View full text Add to dashboard Cite

In this work we study the feasibility of high-bandwidth, secure communications on generic machines equipped with the latest CPUs and General-Purpose Graphical Processing Units (GPGPU). We first analyze the suitability of current Nehalem CPU architectures. We show in particular that high performance CPUs are not sufficient by themselves to reach our performance objectives, and that encryption is the main bottleneck. Therefore we also consider the use of GPGPU, and more particularly we measure the bandwidth of the AES ciphering on CUDA. These tests lead us to the conclusion that finding an appropriate solution is extremely difficult.

show abstract

“…Este escalonamento segue o mesmo mecanismo do Cilk, com roubo de tarefas, e atua no controle de granularidade entre os nós disponíveis à medida que os VT's ficam ociosos. Outros trabalhos em KAAPI [12,5] relatam a eficiência e estabilidade satisfatória do ambiente, no entanto, nenhum estudo avalia a complexidade de código necessária na programac ¸ão de aplicac ¸ões paralelas.…”

Section: Dinamismo E Aplicac ¸õEs Paralelas Irregularesunclassified

Controle de Granularidade com threads em Programas MPI Dinâmicos

Lima

Maillard

2008

Anais Do IX Simpósio Em Sistemas Computacionais De Alto Desempenho (SSCAD 2008)

View full text Add to dashboard Cite

O controle de granularidade é um fator importante no desempenho de programas paralelos. Problemas estáticos adaptam sua granularidade pela decomposição e atribuição de dados a cada tarefa, mas em irregulares não é possível prever a carga de trabalho antes da execução. Os irregulares que utilizam decomposição recursiva, como ordenação, necessitam de dinamismo com suporte a criação de tarefas sob demanda. Alguns ambientes de programação, como Cilk e KAAPI, oferecem dinamismo e trabalham com granularidade através do conceito abstrato de tarefa porém possuem limitações que dificultam seu uso em PAD. O MPI, padrão de fato em PAD, oferece dinamismo de processos e uso de threads mas atribui à implementação especificar o comportamento na criação de uma tarefa. Este trabalho propõe investigar as vantagens no controle de granularidade com threads em programas MPI dinâmicos, através da substituição da criação de processos por tarefas onde um mecanismo (libSpawn) decide entre lançar processo(s) ou thread(s). Os resultados obtidos com o programa de ordenação Cilksort, que segue o modelo Divisão-e-Conquista, demonstram ganhos de até 85% em criação de tarefas e comunicação.

show abstract

Adaptive loops with kaapi on multicore and grid

Cited by 9 publications

References 9 publications

Generic design of Chinese remaindering schemes

Generic design of Chinese remaindering schemes

Parallel arithmetic encryption for high-bandwidth communications on multicore/GPGPU platforms

Controle de Granularidade com threads em Programas MPI Dinâmicos

Contact Info

Product

Resources

About