EZTrace: A Generic Framework for Performance Analysis

Trahay, François; Rue, François; Faverge, Mathieu; Ishikawa, Yutaka; Namyst, Raymond; Dongarra, Jack

doi:10.1109/ccgrid.2011.83

Cited by 31 publications

(19 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the MPI-based benchmarks, we used the eztrace framework [41] to trace all MPI messages sent by tasks and built a communication matrix based on the number of messages sent between tasks. We also generated the communication matrices using the number of bytes and, although the value of each individual cell was different, the overall pattern was the same.…”

Section: Generating the Communication Matricesmentioning

confidence: 99%

EagerMap

Cruz

Diener

Pilla

et al. 2018

ACM Trans. Parallel Comput.

View full text Add to dashboard Cite

HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

show abstract

Section: Generating the Communication Matricesmentioning

confidence: 99%

EagerMap

Cruz

Diener

Pilla

et al. 2018

ACM Trans. Parallel Comput.

View full text Add to dashboard Cite

show abstract

“…2) Generating the Communication Matrices: For the MPI based benchmarks, we used the eztrace framework [17] to trace all MPI messages sent by tasks and built a communication matrix based on the number of messages sent between tasks. For the benchmarks that use shared memory for implicit communication using memory accesses, we built a memory tracer based on the Pin binary analysis tool [18], similarly to [19].…”

Section: A Methodology Of the Mapping Comparisonmentioning

confidence: 99%

An Efficient Algorithm for Communication-Based Task Mapping

Cruz

Diener

Pilla

et al. 2015

2015 23rd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing

View full text Add to dashboard Cite

The communication between tasks of a parallel application is an important characteristic to consider when mapping tasks to computing cores due to possible differences in communication performance. Within a machine, performance differences are introduced by the memory hierarchy, in which cache memories can be shared by groups of cores and intrachip interconnections are faster than inter-chip interconnections. In cluster and grid systems, the network imposes an additional communication latency. By mapping tasks that communicate to cores nearby on the memory hierarchy, or to the same nodes in clusters or grids, the communication of parallel applications is optimized, leading to increased performance and energy efficiency.In the task mapping context, one of the most important aspects to be considered is the mapping algorithm, as it determines the improvements that can be achieved. Since the problem of finding the best mapping is NP-Hard, heuristics must be employed to find an approximate solution in feasible time. In this paper, we present EagerMap, a new algorithm to perform communication-based mapping that is based on a greedy grouping strategy applied hierarchically. Experimental evaluation indicates that the execution time of our algorithm is 10 times faster than the state-of-the-art, and presents higher performance improvements. Due to its low execution time and high stability, EagerMap is also suitable for online task mapping, where tasks are migrated during execution.

show abstract

“…To analyze precisely the behavior of MPI implementations when running the benchmark, we collected execution traces. The traces were generated using the EZTrace framework for performance analysis [21]. When running an application, EZTrace intercepts the calls to a set of functions and records time-stamped events for each call to these functions.…”

Section: B Interleaved Mpi and Infiniband Tracesmentioning

confidence: 99%

MPI Overlap: Benchmark and Analysis

Denis

2016

2016 45th International Conference on Parallel Processing (ICPP)

View full text Add to dashboard Cite

In HPC applications, one of the major overhead compared to sequentiel code, is communication cost. Application programmers often amortize this cost by overlapping communications with computation. To do so, they post a non-blocking MPI request, perform computation, and wait for communication completion, assuming MPI communication will progress in background. In this paper, we propose to measure what really happens when trying to overlap non-blocking point-to-point communications with computation. We explain how background progression works, we describe relevant test cases, we identify challenges for a benchmark, then we propose a benchmark suite to measure how much overlap happen in various cases. We exhibit overlap benchmark results on a wide panel of MPI libraries and hardware platforms. Finally, we classify, analyze, and explain the results using low-level traces to reveal the internal behavior of the MPI library.

show abstract

EZTrace: A Generic Framework for Performance Analysis

Cited by 31 publications

References 4 publications

EagerMap

EagerMap

An Efficient Algorithm for Communication-Based Task Mapping

MPI Overlap: Benchmark and Analysis

Contact Info

Product

Resources

About