Reproducible Measurements of MPI Performance Characteristics
Proceedings of the 6th European PVM/MPI Users' Group Meeting on Recent Advances in Parallel Virtual Machine and Message Passing Interface
On Benchmarking Collective MPI Operations
Proceedings of the 9th European PVM/MPI Users' Group Meeting on Recent Advances in Parallel Virtual Machine and Message Passing Interface
Accurately Measuring MPI Broadcasts in a Computational Grid
HPDC '99 Proceedings of the 8th IEEE International Symposium on High Performance Distributed Computing
ICPADS '06 Proceedings of the 12th International Conference on Parallel and Distributed Systems - Volume 2
Building the communication performance model of heterogeneous clusters based on a switched network
CLUSTER '07 Proceedings of the 2007 IEEE International Conference on Cluster Computing
Optimization of collective communications in HeteroMPI
PVM/MPI'07 Proceedings of the 14th European conference on Recent Advances in Parallel Virtual Machine and Message Passing Interface
A Software Tool for Accurate Estimation of Parameters of Heterogeneous Communication Models
Proceedings of the 15th European PVM/MPI Users' Group Meeting on Recent Advances in Parallel Virtual Machine and Message Passing Interface
Accurate Heterogeneous Communication Models and a Software Tool for Their Efficient Estimation
International Journal of High Performance Computing Applications
The analysis of cluster interconnect with the network tests2 toolkit
EuroMPI'11 Proceedings of the 18th European MPI Users' Group conference on Recent advances in the message passing interface
mpicroscope: towards an MPI benchmark tool for performance guideline verification
EuroMPI'12 Proceedings of the 19th European conference on Recent Advances in the Message Passing Interface
Hi-index | 0.00 |
In this paper, we analyze existing MPI benchmarking suites, focusing on two restrictions that prevent them from a wider use in applications and programming systems. The first is a single method of measurement of the execution time of MPI communications implemented by each of the suites. The second one is the design of the suites in the form of a standalone executable program that cannot be easily integrated into applications or programming systems. We present a more flexible benchmarking package, MPIBlib, that provides multiple methods of measurement, both operation-independent and operation-specific. This package can be used not only for benchmarking but also as a library in applications and programming systems for communication performance modeling and optimization of MPI operations.