Characterization of MPI communication primitives on a heterogeneous cluster / Siti Arpah Ahmad ... [et al.]

This paper focuses on the performance of basic communication primitives, namely the overlap of message transfer with computation in the point-to-point communication within a small cluster of four nodes. The mpptest has been implemented to measure the basic performance of MPI message passing routines...

Full description

Saved in:
Bibliographic Details
Main Authors: Ahmad, Siti Arpah (Author), Mohamed Said, Mohamed Faidz (Author), Mohamed Ramli, Norazan (Author), Taib, Mohd Nasir (Author)
Format: Book
Published: Research Management Institute (RMI), 2009.
Subjects:
Online Access:Link Metadata
Tags: Add Tag
No Tags, Be the first to tag this record!

MARC

LEADER 00000 am a22000003u 4500
001 repouitm_12924
042 |a dc 
100 1 0 |a Ahmad, Siti Arpah  |e author 
700 1 0 |a Mohamed Said, Mohamed Faidz  |e author 
700 1 0 |a Mohamed Ramli, Norazan  |e author 
700 1 0 |a Taib, Mohd Nasir  |e author 
245 0 0 |a Characterization of MPI communication primitives on a heterogeneous cluster / Siti Arpah Ahmad ... [et al.] 
260 |b Research Management Institute (RMI),   |c 2009. 
500 |a https://ir.uitm.edu.my/id/eprint/12924/1/AJ_SITI%20ARPAH%20AHMAD%20SRJ%2009%201.pdf 
520 |a This paper focuses on the performance of basic communication primitives, namely the overlap of message transfer with computation in the point-to-point communication within a small cluster of four nodes. The mpptest has been implemented to measure the basic performance of MPI message passing routines with a variety of message sizes. The mpptest is capable of measuring performance with many participating processes thus exposing contention and scalability problems. This enables programmers to select message sizes in order to isolate and evaluate sudden changes in performance. Investigating these matters is interesting in that non-blocking calls have the advantage of allowing the system to schedule communications even when many processes are running simultaneously. On the other hand, understanding the characteristics of computation and communication overlap is significant, because high performance kernels often strive to achieve this, since it is both advantageous with respect to data transfer and latency hiding. The results indicate that certain overlap sizes utilize greater node processing power either in blocking send and receive operations or non-blocking send and receive operations. The results have elucidated a detailed MPI characterization of the performance regarding the overlap of message transfer with computation in a small cluster system. 
546 |a en 
690 |a Parallel processing (Electronic computers) 
655 7 |a Article  |2 local 
655 7 |a PeerReviewed  |2 local 
787 0 |n https://ir.uitm.edu.my/id/eprint/12924/ 
787 0 |n https://srj.uitm.edu.my/ 
856 4 1 |u https://ir.uitm.edu.my/id/eprint/12924/  |z Link Metadata