Parallel Virtual File System 1.6.3 review
DownloadThe goal of the Parallel Virtual File System (PVFS) Project is to explore the design, implementation, and uses of parallel I/O
|
|
The goal of the Parallel Virtual File System (PVFS) Project is to explore the design, implementation, and uses of parallel I/O. PVFS serves as both a platform for parallel I/O research as well as a production file system for the cluster computing community. PVFS is currently targeted at clusters of workstations, or Beowulfs.
The PVFS project is conducted jointly between The Parallel Architecture Research Laboratory at Clemson University and The Mathematics and Computer Science Division at Argonne National Laboratory.
Additional funding for the PVFS project comes from NASA Goddard Space Flight Center Code 930 and The National Computational Science Alliance through the National Science Foundation's Partnerships for Advanced Computational Infrastructure.
Here are some key features of "Parallel Virtual File System":
Compatibility with existing binaries
Ease of installation
User-controlled striping of files across nodes
Multiple interfaces, including a MPI-IO interface via ROMIO
Utilizes commodity network and storage hardware
PVFS supports the UNIX I/O interface and allows existing UNIX I/O programs to use PVFS files without recompiling. The familiar UNIX file tools (ls, cp, rm, etc.) will all operate on PVFS files and directories as well. This is accomplished via a Linux kernel module which is provided as a separate package.
PVFS is easy to install. The Quick Start page describes how to set up a simple installation. Scripts and test applications are included to help with configuration, testing for correct operation, and performance evaluation.
PVFS stripes file data across multiple disks in different nodes in a cluster. By spreading out file data in this manner, larger files can be created, potential bandwidth is increased, and network bottlenecks are minimized. A 64-bit interface is implemented as well, allowing large (more than 2GB) files to be created and accessed.
What's New in This Release:
fixes to build under redhat 2.4.20-20.9 kernel
checks for missing headers on ancient 2.4 kernels
removal of unused fields from metadata and control messages
updated pvfsd rc file
handle various combinations of redhat kernel patches
fixes to kpvfsd that had previously been fixed in user space
fix for sftp directory reading problems
fix to avoid oopses on clients when bad errno values returned
adjustments to deal wiht gcc 3.4.0 issues
fixes to work with redhat 9
fixes to compile kernel code for xx86_64
updated logging intrastructure
fixes to error reporting
update of makefile in pvfs core
fix to bug in pvfs_open.c where pcount < 0 caused errors
removed duplicated code in mgr
fix for problem with bvrecv() optimization
more unused code removal
adjustments to keep portland group compiler happy
removed sd_path from metadata
new iod.rc and mgr.rc files
fixes to chmod
fixes to get correct fs_ino back to clients
multiple mtime fixes
streamlined pvfs file detection
shared library building
fixed closing socket bug in iod
reordered messaging in mgr to iods to improve concurrency
rename fix
improved metadata file checking
do_access permission fixes
fix for "iod out of space" client hang
cleaner dead socket handling, including special handling of case where mgr would close fds that clients were using
noninteractive mkmgrconf
Parallel Virtual File System 1.6.3 keywords