|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] RE: Why FCP doesn't need RDMA? It has a better way.David, This example is of .5MB transfers using jumbo frames. It would be difficult to extrapolate performance as related to normal SCSI use. It does suggest that such hardware able to implement scatter/gather is useful. It would be equally useful for any protocol. Doug > -----Original Message----- > From: owner-ips@ece.cmu.edu [mailto:owner-ips@ece.cmu.edu]On Behalf Of > David Robinson > Sent: Thursday, September 28, 2000 11:36 AM > To: ips@ece.cmu.edu > Subject: RE: Why FCP doesn't need RDMA? It has a better way. > > > > From an iSCSI viewpoint, I don't see iSCSI by itself as being sufficient > > to motivate a protocol-independent RDMA - an iSCSI HBA could understand > > the iSCSI headers and interact with DMA in the same fashion as existing > > HBAs. The task before those interested in RDMA is to identify a set > > of protocols for which a common RDMA mechanism makes sense from > > an engineering standpoint. I tend to agree with the previous emails > > that iSCSI could make optional use of a common RDMA mechanism > > if available, but must not REQUIRE its use. > > As another data point to show that RDMA is not required, a recently > reported group acheiving 980+Mbps on a gigabit ethernet using > NFS. NFS is not nearly as simple to parse headers as FCP or > the current iSCSI draft, and the authors were able to reprogram > an Altheon card's firmware to acheive it. I suspect a good design > engineer could do the same for iSCSI without RDMA. > > http://people.freebsd.org/~ken/zero_copy/ > > So while RDMA is interesting, I see no reason to REQUIRE it for iSCSI. > > -David > >
Home Last updated: Tue Sep 04 01:07:00 2001 6315 messages in chronological order |