I have implemented a DLM for locking at the byte-range level in the version-lock-actual-branch. A link to the relevant paper is
http://cholera.ece.northwestern.edu/~aching/research_webpage/publications/20 07sc_2.pdf You would also need my MPI-IO PVFS2 driver to support MPI-IO atomic mode. I have not integrated these locks with the PVFS kernel interface (although it would be an important building block for consistent client side caching in PVFS). If you are interested in this, feel free to contact me. Avery On 1/4/08 8:01 AM, "Robert Latham" <[EMAIL PROTECTED]> wrote: > On Fri, Jan 04, 2008 at 12:01:50AM -0800, Murali Vilayannur wrote: >> Craig, >> >>> I know that PVFS2 does not support locking and understand the >>> benefits of this design. However, I have a question about what >>> happens to file writes for contiguous, non-overlapping writes. >> >> I think Avery has patches to support locking at the file/byte range level. >> I am not sure if they are in some private branch though and it is >> likely they may >> need some additional work to integrate with the VFS...Avery, Rob? > > for the curious, that work lives in 'version-lock-actual-branch' > > Thanks for the detailed answer, Murali! > ==rob _______________________________________________ Pvfs2-users mailing list [email protected] http://www.beowulf-underground.org/mailman/listinfo/pvfs2-users
