FC, FCoE, iSCSI, HyperSCSI, etc…

Just finished reading Mark Lewis’ latest blog in which Mark talks about the evolution of Fibre Channel to FCoE (Fibre Channel over Ethernet).  I have to agree that the convergence at the network layer will continue and the maturity and pervasiveness of Ethernet will surely prevail over FC (fibre channel).  Interestingly enough HyperSCSI or SCSI-over-Ethernet did not prevail over iSCSI.  This could be due to the fact that there was a FC alternative.  I have a theory that with the exception of the largest of shops most organizations are looking to converge and consolidate network infrastructure.  Most of these shops are looking to leverage functionally available in Layer 3 and Layer 4, while there is a obvious performance benefit of a direct storage connection at Layer 2 for many shops this does not outweigh their familiarity and comfort with the functionality provided at Layer 3 and 4.  For example, IPsec or OSPF – The Fibre Channel community has developed equivalents to some of the functionality that we see in the IP world, for example OSPF in the  Fibre Channel world is FSPF (Fabric Shortest Path First).  The big question in my mind is do we continue to build develop functionality at the transport and network layer or does the storage industry focus on the transport layer (Layer 4) and let the networking industry focus on the network layer (Layer 3), data link layer (Layer 2) and the physical layer (Layer 1) seems to make more sense to me from a long term perspective.  Not sure why the storage industry should be reinventing OSPF, RIP, EIGRP, IPsec, etc…but that is just my opinion – but based on the recent adoption rate of iSCSI I think it might be the opinion of others as well.

One thought on “FC, FCoE, iSCSI, HyperSCSI, etc…

Leave a Reply to Jospeh Buza Cancel reply

Your email address will not be published. Required fields are marked *