Implementing MPI-IO atomic mode and shared file pointers using MPI one-sided communication

被引:0
作者
Latham, Robert [1 ]
Ross, Robert [1 ]
Thakur, Rajeev [1 ]
机构
[1] Argonne Natl Lab, Div Math & Comp Sci, Argonne, IL 60439 USA
关键词
MPI-IO; RMA; message-passing; MPI-2; implementation;
D O I
10.1177/1094342007077859
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
The ROMIO implementation of the MPI-IO standard provides a portable infrastructure for use on top of a variety of underlying storage targets. These targets vary widely in their capabilities, and in some cases additional effort is needed within ROMIO to support all MPI-IO semantics. Two aspects of the interface that can be problematic to implement are MPI-IO atomic mode and the shared file pointer access routines. Atomic mode requires enforcing strict consistency semantics, and shared file pointer routines require communication and coordination in order to atomically update a shared resource. For some file systems, native locks may be used to implement these features, but not all file systems have lock support. In this work, we describe algorithms for implementing efficient mutex locks using MPI-1 and the one-sided capabilities from MPI-2. We then show how these algorithms may be used to implement both MPI-IO atomic mode and shared file pointer methods for ROMIO without requiring any features from the underlying file system. We show that these algorithms can outperform traditional file system lock approaches. Because of the portable nature of these algorithms, they are likely useful in a variety of situations where distributed locking or coordination is needed in the MPI-2 environment.
引用
收藏
页码:132 / 143
页数:12
相关论文
共 20 条
[1]   Data management and transfer in high-performance computational grid environments [J].
Allcock, B ;
Bester, J ;
Bresnahan, J ;
Chervenak, AL ;
Foster, I ;
Kesselman, C ;
Meder, S ;
Nefedova, V ;
Quesnel, D ;
Tuecke, S .
PARALLEL COMPUTING, 2002, 28 (05) :749-771
[2]  
Carns PH, 2000, USENIX ASSOCIATION PROCEEDINGS OF THE 4TH ANNUAL LINUX SHOWCASE AND CONFERENCE, ATLANTA, P317
[3]  
CORBETT PF, 1994, PROCEEDINGS OF THE SCALABLE HIGH-PERFORMANCE COMPUTING CONFERENCE, P63, DOI 10.1109/SHPCC.1994.296628
[4]   SPIFFI - A scalable parallel file system for the intel paragon [J].
Freedman, CS ;
Burger, J ;
DeWitt, DJ .
IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 1996, 7 (11) :1185-1200
[5]  
Gropp WaT, 1999, USING MPI 2 ADV FEAT
[6]   ALGORITHMS FOR SCALABLE SYNCHRONIZATION ON SHARED-MEMORY MULTIPROCESSORS [J].
MELLORCRUMMEY, JM ;
SCOTT, ML .
ACM TRANSACTIONS ON COMPUTER SYSTEMS, 1991, 9 (01) :21-65
[7]  
MPI Forum, 1997, MPI 2 EXT MESS PASS
[8]  
Pierce P., 1989, Proceedings of the Fourth Conference on Hypercubes, Concurrent Computers and Applications, P155
[9]  
Thakur R, 2005, LECT NOTES COMPUT SC, V3666, P119
[10]  
Thakur R., 1999, P 6 WORKSH I O PAR D, P23, DOI DOI 10.1145/301816.301826