Automatic Transformation for Overlapping Communication and Computation

Hu, Changjun; Shao, Yewei; Wang, Jue; Li, Jianjiang

doi:10.1007/978-3-540-88140-7_19

Cited by 3 publications

(2 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There has also been significant work undertaken looking at optimising communications in MPI programs. A number of authors have looked at compiler based optimisations to provide automatic overlapping of communications and computation in existing parallel programs [10,6,11]. These approaches have shown that performance improvements can be obtained, generally evaluated against kernel benchmarks such as the NAS parallel benchmarks, by transforming user specified blocking communication code to non-blocking communication functionality, and using static compiler analysis to determine where the communications can be started and finished.…”

Section: Related Workmentioning

confidence: 99%

MDMP: Managed Data Message Passing

Jackson¹,

Strand

2013

2013 15th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing

View full text Add to dashboard Cite

MDMP is a new parallel programming approach that aims to provide users with an easy way to add parallelism to programs, optimise the message passing costs of traditional scientific simulation algorithms, and enable existing MPIbased parallel programs to be optimised and extended without requiring the whole code to be re-written from scratch. MDMP utilises a directives based approach to enable users to specify what communications should take place in the code, and then implements those communications for the user in an optimal manner using both the information provided by the user and data collected from instrumenting the code and gathering information on the data to be communicated. In this paper we present the basic concepts and functionality of MDMP and discuss the performance that can be achieved using our prototype implementation of MDMP on some simple benchmark cases.

show abstract

Section: Related Workmentioning

confidence: 99%

MDMP: Managed Data Message Passing

Jackson¹,

Strand

2013

2013 15th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing

View full text Add to dashboard Cite

show abstract

“…Other research has explored compiler-based MPI code optimizations [7], [8], or acceleration of MPI point-topoint communication [9]. Many approaches use code motion techniques to increase communication/computation overlap, an orthogonal technique that we could easily combine with our approach.…”

Section: Related Workmentioning

confidence: 99%

Exploitation of Dynamic Communication Patterns through Static Analysis

Preissl

Supinski

Schulz

et al. 2010

2010 39th International Conference on Parallel Processing

View full text Add to dashboard Cite

Abstract-Collective operations can have a large impact on the performance of parallel applications. However, the ideal implementation of a particular collective communication often depends on both the application and the targeted machine structure. Our approach combines dynamic and static analysis techniques to identify common collective communication patterns expressed as point-to-point calls and transforms them into equivalent MPI collectives. We first detect potential collective communication patterns in runtime traces and associate them with the corresponding source code regions. If our static analysis verifies that the introduction of collectives is safe for any program flow, we then replace the original communication primitives with their collective counterpart. In this paper we introduce the necessary algorithms to determine the safety of these transformations and we demonstrate several use cases, including automatic use of new extensions to the MPI standard such as nonblocking collective operations. The use of dynamic analysis significantly reduces compile times, resulting in a speed-up of about 50 for source transformations of HPL due to more directed analysis capabilities and also dramatically decreases complexity of the underlying static analysis.

show abstract

Compiler-Assisted Overlapping of Communication and Computation in MPI Applications

Guo

Meng

et al. 2016

2016 IEEE International Conference on Cluster Computing (CLUSTER)

View full text Add to dashboard Cite

Automatic Transformation for Overlapping Communication and Computation

Cited by 3 publications

References 17 publications

MDMP: Managed Data Message Passing

MDMP: Managed Data Message Passing

Exploitation of Dynamic Communication Patterns through Static Analysis

Compiler-Assisted Overlapping of Communication and Computation in MPI Applications

Contact Info

Product

Resources

About