Analysis of the MPI-IO optimization levels with the PIOViz Jumpshot enhancement

Authors:
Thomas Ludwig;Stephan Krempel;Michael Kuhn;Julian Kunkel;Christian Lohse
Affiliations:
Ruprecht-Karls-Universität Heidelberg, Heidelberg, Germany;Ruprecht-Karls-Universität Heidelberg, Heidelberg, Germany;Ruprecht-Karls-Universität Heidelberg, Heidelberg, Germany;Ruprecht-Karls-Universität Heidelberg, Heidelberg, Germany;Ruprecht-Karls-Universität Heidelberg, Heidelberg, Germany
Venue:
PVM/MPI'07 Proceedings of the 14th European conference on Recent Advances in Parallel Virtual Machine and Message Passing Interface
Year:
2007

Citing 5
Cited 3

Using MPI-2: Advanced Features of the Message Passing Interface

Using MPI-2: Advanced Features of the Message Passing Interface
The Paradyn Parallel Performance Measurement Tool

Computer
The Tau Parallel Performance System

International Journal of High Performance Computing Applications
Performance Evaluation of the PVFS2 Architecture

PDP '07 Proceedings of the 15th Euromicro International Conference on Parallel, Distributed and Network-Based Processing
Tracing the MPI-IO calls' disk accesses

EuroPVM/MPI'06 Proceedings of the 13th European PVM/MPI User's Group conference on Recent advances in parallel virtual machine and message passing interface

Directory-Based Metadata Optimizations for Small Files in PVFS

Euro-Par '08 Proceedings of the 14th international Euro-Par conference on Parallel Processing
Bottleneck Detection in Parallel File Systems with Trace-Based Performance Monitoring

Euro-Par '08 Proceedings of the 14th international Euro-Par conference on Parallel Processing
Visual analysis of I/O system behavior for high-end computing

Proceedings of the third international workshop on Large-scale system and application performance

Quantified Score

Hi-index	0.00

Visualization

Abstract

With MPI-IO we see various alternatives for programming file I/O. The overall program performance depends on many different factors. A new trace analysis environment provides deeper insight into the client/server behavior and visualizes events of both process types. We investigate the influence of making independent vs. collective calls together with access to contiguous and non-contiguous data regions in our MPI-IO program. Combined client and server traces exhibit reasons for observed I/O performance.