A Simulation Framework to Automatically Analyze the Communication-Computation Overlap in Scientific Applications

Subotic, Vladimir; Sancho, José Carlos; Labarta, Jesús; Valero, Mateo

doi:10.1109/cluster.2010.33

Cited by 9 publications

(5 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sancho et al [14] provided a theoretical estimation of the overlapping potential in scientific codes by modeling a real application with one iterative loop and parameters that roughly describe the computation pattern. Continuing this work, in our former study [16], we designed an automatic simulation environment that estimates applications' overlapping potential and showed that legacy real MPI applications have significant potential for overlap. Further on, in this paper we try to inspect in more detail the influence on the overlapping potential.…”

Section: Related Workmentioning

confidence: 99%

The Impact of Application's Micro-Imbalance on the Communication-Computation Overlap

Subotic

Sancho

Labarta

et al. 2011

2011 19th International Euromicro Conference on Parallel, Distributed and Network-Based Processing

Self Cite

View full text Add to dashboard Cite

Although the community sees overlapping communication and computation as a perspective avenue for advancing parallel execution, it remains unclear what type of applications, under which conditions, and to which extent could benefit from this technique. To tackle this issue, we designed a simulation environment that allowed us to profoundly study overlap. We found out that overlapping potential in an application is determined by the application's parallel behavior and the pattern by which each process locally produces/consumes data involved in communication. We identified two behaviors that directly influence the application's overlapping potential -we name them microscopic imbalance of computation and microscopic imbalance of communication.In an application that expresses some of these two behaviors, a fine-grain overlapping technique can achieve a significant execution speedup, a speedup that can even be higher than 2. We believe that our findings can help a programmer estimate how much his application could benefit from overlap, and therefore decide whether implementing that technique is worth the effort.

show abstract

Section: Related Workmentioning

confidence: 99%

The Impact of Application's Micro-Imbalance on the Communication-Computation Overlap

Subotic

Sancho

Labarta

et al. 2011

2011 19th International Euromicro Conference on Parallel, Distributed and Network-Based Processing

Self Cite

View full text Add to dashboard Cite

show abstract

“…Especially for network performance, only limited options are available. The simulator Dimemas [19,41] can be used to analyze the performance of message-passing applications. Its traces can be further analyzed with the visualization tool Paraver [26].…”

Section: Related Workmentioning

confidence: 99%

Automated Network Performance Characterization for HPC Systems

Bartelheimer,

Zhu,

Neuwirth

2024

IJNC

View full text Add to dashboard Cite

In this work, we propose the Network Performance Collector (NPC) workflow for automated network performance characterization. The workflow is based on the collection, processing as well as visualization of network performance metrics such as throughput and latency and can be used for analysis with different network performance models. Depending on the chosen model, benchmark tools such as iperf or sockperf as well as microbenchmarks specific to parallel programming models can be automated and orchestrated for data collection with the NPC. The data obtained can then be used by NPC, for example, to validate and characterize the performance of the underlying network or to analyze the system boundaries for a particular application. We provide a prototype implementation of the proposed workflow and demonstrate its effectiveness by automating an extended Roofline model analysis.

show abstract

“…On a heterogeneous environment, overlapping communication and computation between tasks is difficult. The performance of MPI calls is limited for the overlapping of dynamic applications as highlighted in [11]. A parallel phase model approach is introduced to improve the performance of irregular memory access application by providing effective overlap of communication and computation [12].…”

Section: A Message Passing Interface (Mpi)mentioning

confidence: 99%

Performance enhancing factors for manycore architectures: State-of-the-art

Tamizharasan

Yadav

Ramasubramanian

et al. 2014

2014 First International Conference on Networks &Amp; Soft Computing (ICNSC2014)

View full text Add to dashboard Cite

Manycore architecture system includes more number of processing elements to improve the performance while sustaining power considerations. Accelerating heterogeneous manycore computing elements involves huge amount of memory copy, computation and thread management. Applications of manycore architectures range from desktop computer to warehouse-scale computer. In this paper, the state-of-the-art trends and techniques of few manycore architectures that address various memory and performance issues are presented. Different parallel programming models that address issues on thread management, accessing virtual shared memory, dynamic load balancing, memory hierarchy and inter-language compatibility are discussed among some of the manycore architectures like NVIDIA's Graphics Processing Units (GPU), Intel's Many Integrated Core (MIC) and a joint venture of AMD's Accelerated Processing Units (APU).

show abstract

A Simulation Framework to Automatically Analyze the Communication-Computation Overlap in Scientific Applications

Cited by 9 publications

References 16 publications

The Impact of Application's Micro-Imbalance on the Communication-Computation Overlap

The Impact of Application's Micro-Imbalance on the Communication-Computation Overlap

Automated Network Performance Characterization for HPC Systems

Performance enhancing factors for manycore architectures: State-of-the-art

Contact Info

Product

Resources

About