Messages in this thread | | | Date | Tue, 30 Nov 1999 16:55:52 +0000 (GMT) | From | "Mark O'Neill" <> | Subject | Help with non-standard mod to NFS Kernel daemon ... |
| |
I'm not sure if I am writing to the right people -- but I guess I have to start somewhere! I have been writing a LINUX based system for implementing persistent (e.g. long running) parallel computations to support bioinformatics projects in Oxford, Newcastle and at the Natural History Museum in the UK. I have come up with a system which is fault tolerant, provides a couple of parallel computing paradigms and which will run (in theory on any, and in practice at least 5!) POSIX.1b conformant systems.
One of the nicest aspects of this system is a shared heap model which permits processes to use mmap() (as opposed to sbrk() to create and share data heaps. The problem is while this works very nicely if the co-operating processes are on the same machine, it fails under NFS because user processes on a client machine cannot tell the NFS kernel daemon on that machine to flush all cached data back to the physical file on the NFS server. How hard would it be to implement an additional system call [say nfssync()] to perform this function? Typically this would not impinge on NFS filesystem performance, as this (slow) operation would only need to be called when a process has finished modifying a shared heap in order to provide up-to-date data for other clients of that heap.
Any help would be appreciated. I am quite happy to add this modification to the kernel myself, given sufficient guidance!
FYI you can see the sort distributed systems we have been building, and a few example applications at the PUPS Web Site http://fabre.oum.ox.ac.uk
Regards
Mark O'Neill
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |