I was thinking that a "virtual san" would primarily be for the local virt ual machines and if implemented as a new flavor of SFS then the cross-system use would just be icing on the cake. What I don't like about NFS is the numbe r of hops the data has to take even when all of the processes are in the sa me physical platform (in this sense LPARs are different platforms).
DASD -> CP -> SFS -> NFS -> TCPIP -> communications implementation -> Lin ux And I don't know how many hops the data takes inside SFS, NFS or Linux. T he The kind of path I was thinking a "virtual san" in a single platform migh t have is: DASD -> CP -> SFS -> CP -> LINUX I understand that extending the connectivity beyond a single platform wil l add hops to the path but that is part of the price you pay. When I REALLY need to extent data from one of my linux servers to beyond the platform t hen NFS is reasonable and is the price you pay. /Tom Kern On Wed, 24 Jan 2007 09:46:33 +0100, Rob van der Heij <[EMAIL PROTECTED]> wrote: >On 1/23/07, Thomas Kern <[EMAIL PROTECTED]> wrote: >> Could this "Virtual San" be some modification of the Shared/Byte File >> System server? With IPGATE, that could even be used across LPARs via >> hypersockets and across physical machines via other TCPIP connections. > >Well, that's basically NFS to serve BFS space. While I admit we never >had the need to study it in more detail, so far I have not been >impressed by the performance of the VM NFS server. If you're going NFS >then I suppose it would be more realistic to have a Linux server host >it. But each Linux virtual machine would still need its own local disk >space to hold the basic materials to get the network going (or maybe >do something with an NSS or implement booting from network). >Have not looked at iSCSI in detail yet. Some folks I respect tell me >not to do that... > ...snipped...