TY - GEN
T1 - Implementing MPI-IO atomic mode without file system support
AU - Ross, Robert
AU - Latham, Robert
AU - Gropp, William D
AU - Thakur, Rajeev
AU - Toonen, Brian
PY - 2005
Y1 - 2005
N2 - The ROMIO implementation of the MPI-IO standard provides a portable infrastructure for use on top of any number of different underlying storage targets. These different targets vary widely in their capabilities, and in some cases, additional effort is needed within ROMIO to support the complete MPI-IO semantics. One aspect of the interface that can be problematic to implement is the MPI-IO atomic mode. This mode requires enforcing strict consistency semantics. For some file systems, native locks may be used to enforce these semantics, but not all file systems have lock support. In this work, we describe two algorithms for implementing efficient mutex locks using MPI-1 and MPI-2 capabilities. We then show how these algorithms may be used to implement a portable MPI-IO atomic mode for ROMIO. We evaluate the performance of these algorithms and show that they impose little additional overhead on the system. Because of the low-overhead nature of these algorithms, they are likely useful in a variety of situations where distributed locks are needed in the MPI-2 environment.
AB - The ROMIO implementation of the MPI-IO standard provides a portable infrastructure for use on top of any number of different underlying storage targets. These different targets vary widely in their capabilities, and in some cases, additional effort is needed within ROMIO to support the complete MPI-IO semantics. One aspect of the interface that can be problematic to implement is the MPI-IO atomic mode. This mode requires enforcing strict consistency semantics. For some file systems, native locks may be used to enforce these semantics, but not all file systems have lock support. In this work, we describe two algorithms for implementing efficient mutex locks using MPI-1 and MPI-2 capabilities. We then show how these algorithms may be used to implement a portable MPI-IO atomic mode for ROMIO. We evaluate the performance of these algorithms and show that they impose little additional overhead on the system. Because of the low-overhead nature of these algorithms, they are likely useful in a variety of situations where distributed locks are needed in the MPI-2 environment.
UR - http://www.scopus.com/inward/record.url?scp=33845289423&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=33845289423&partnerID=8YFLogxK
U2 - 10.1109/CCGRID.2005.1558687
DO - 10.1109/CCGRID.2005.1558687
M3 - Conference contribution
AN - SCOPUS:33845289423
SN - 0780390741
SN - 9780780390744
T3 - 2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005
SP - 1135
EP - 1142
BT - 2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005
T2 - 2005 IEEE International Symposium on Cluster Computing and the Grid, CCGrid 2005
Y2 - 9 May 2005 through 12 May 2005
ER -