Abstract
Many scientific applications use parallel I/O to meet the low latency and high bandwidth I/O requirement. Among many available parallel I/O operations, collective I/O is one of the most popular methods when the storage layouts and access patterns of data do not match. The implementation of collective I/O typically involves disk I/O operations followed by interprocessor communications. Also, in many I/O-intensive applications, parallel I/O operations are usually followed by parallel computations. This paper presents a comparative study of different overlap strategies in parallel applications. We have experimented with four different overlap strategies 1) Overlapping I/O and communication; 2) Overlapping I/O and computation; 3) Overlapping computation and communication; and 4) Overlapping I/O, communication, and computation. All experiments have been conducted on a Linux Cluster and the performance results obtained are very encouraging. On an average, we have enhanced the performance of a generic collective read call by 38%, the MxM benchmark by 26%, and the FFT benchmark by 34%.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 43-49 |
| Number of pages | 7 |
| Journal | Operating Systems Review (ACM) |
| Volume | 42 |
| Issue number | 6 |
| DOIs | |
| State | Published - Oct 1 2008 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Information Systems
- Hardware and Architecture
- Computer Networks and Communications
Cite this
}
Comparative evaluation of overlap strategies with study of I/O overlap in MPI-IO. / Patrick, Christina M.; Son, Seungwoo; Kandemir, Mahmut.
In: Operating Systems Review (ACM), Vol. 42, No. 6, 01.10.2008, p. 43-49.Research output: Contribution to journal › Article
TY - JOUR
T1 - Comparative evaluation of overlap strategies with study of I/O overlap in MPI-IO
AU - Patrick, Christina M.
AU - Son, Seungwoo
AU - Kandemir, Mahmut
PY - 2008/10/1
Y1 - 2008/10/1
N2 - Many scientific applications use parallel I/O to meet the low latency and high bandwidth I/O requirement. Among many available parallel I/O operations, collective I/O is one of the most popular methods when the storage layouts and access patterns of data do not match. The implementation of collective I/O typically involves disk I/O operations followed by interprocessor communications. Also, in many I/O-intensive applications, parallel I/O operations are usually followed by parallel computations. This paper presents a comparative study of different overlap strategies in parallel applications. We have experimented with four different overlap strategies 1) Overlapping I/O and communication; 2) Overlapping I/O and computation; 3) Overlapping computation and communication; and 4) Overlapping I/O, communication, and computation. All experiments have been conducted on a Linux Cluster and the performance results obtained are very encouraging. On an average, we have enhanced the performance of a generic collective read call by 38%, the MxM benchmark by 26%, and the FFT benchmark by 34%.
AB - Many scientific applications use parallel I/O to meet the low latency and high bandwidth I/O requirement. Among many available parallel I/O operations, collective I/O is one of the most popular methods when the storage layouts and access patterns of data do not match. The implementation of collective I/O typically involves disk I/O operations followed by interprocessor communications. Also, in many I/O-intensive applications, parallel I/O operations are usually followed by parallel computations. This paper presents a comparative study of different overlap strategies in parallel applications. We have experimented with four different overlap strategies 1) Overlapping I/O and communication; 2) Overlapping I/O and computation; 3) Overlapping computation and communication; and 4) Overlapping I/O, communication, and computation. All experiments have been conducted on a Linux Cluster and the performance results obtained are very encouraging. On an average, we have enhanced the performance of a generic collective read call by 38%, the MxM benchmark by 26%, and the FFT benchmark by 34%.
UR - http://www.scopus.com/inward/record.url?scp=70449684261&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=70449684261&partnerID=8YFLogxK
U2 - 10.1145/1453775.1453784
DO - 10.1145/1453775.1453784
M3 - Article
VL - 42
SP - 43
EP - 49
JO - Operating Systems Review (ACM)
JF - Operating Systems Review (ACM)
SN - 0163-5980
IS - 6
ER -