Paper Info
Title
Parallel netCDF: A Scientific High-Performance I/O Interface
Abstract
Dataset storage, exchange, and access play a critical role in scientific applications. For such purposes netCDF serves as a portable and efficient file format and program- ming interface, which is popular in numerous scientific ap- plication domains. However, the original interface does not provide a efficient mechanism for parallel data storage and access. In this work, we present a new parallel interface for writ- ing and reading netCDF datasets. This interface is de- rived with minimal changes from the serial netCDF inter- face but defines semantics for parallel access and is tai- lored for high performance. The underlying parallel I/O is achieved through MPI-IO, allowing for dramatic perfor- mance gains through the use of collective I/O optimizations. We compare the implementationstrategies with HDF5 and analyze both. Our tests indicate programming convenience and significant I/O performanceimprovement with this par- allel netCDF interface.
Year
Venue
Keywords
2003
Clinical Orthopaedics and Related Research
cluster computing,data storage
Field
DocType
Volume
File format,Hierarchical Data Format,Computer science,Computer data storage,Input/output,NetCDF,Parallel port,Semantics,Performance improvement,Distributed computing
Journal
cs.DC/0306
Citations 
PageRank 
References 
15
2.01
6
Authors
7
Name
Order
Citations
PageRank
Jianwei Li117611.56
Wei-keng Liao2109587.98
Alok N. Choudhary33441326.32
Robert Ross42717173.13
Rajeev Thakur53773251.09
William D. Gropp65547548.31
Robert Latham736526.39