iSCSI and FCoE – Which Will Win?

Tags: iSCSI, FCoE, DCB, Data Center Ethernet, Ethernet Storage

Part 1 of 2

When new technologies are in development we often see articles by analysts and media pundits predicting wholesale change, as the emerging new technology enters the market.

This is often portrayed as a head-to-head battle with an existing technology – a fight to the death. 

Unfortunately, this impression is usually the result of briefings by vendors with a big stake in the new technology, without much consideration of the IT realities.  It shouldn’t be surprising then that in the real world, technology uptake rarely happens that way.  In most cases, the new technology coexists with other technologies for an extended period of time, as each finds its place in the market.

A case in point is the forthcoming emergence of Fibre Channel over Ethernet (FCoE), which has already been written about as an iSCSI killer, which prompted an interesting response.

iSCSI deployments to date have typically been in support of Windows or Virtual Server environments:  in tier 2 and tier 3 data centers in large organizations; in the core data center of small/medium enterprises;  and in remote offices.  In other words, iSCSI is usually chosen for environments that Fibre Channel has had difficulty penetrating due to cost, complexity, functionality and support issues.

The availability of FCoE -- which, remember, is being specifically designed to maintain backward compatibility with existing Fibre Channel endpoint infrastructure (which means World Wide Names, Fibre Channel fabric management, “converged” host bus adapters, and Fibre Channel support matrix complexities) -- seems very unlikely to replace or displace continued iSCSI growth in these sweet spot environments.

All of this leaves the question of why and where FCoE will be deployed. The “why” part of the question is fairly simple – the fundamental decision criterion for FCoE is a desire to move to a single network fabric in the data center for storage and data communications. 

I’ll give you my view on the “where” in my next post.  Stay tuned.

Powered by Qumana

Comments

[Stephen Foskett]

I think you and I agree on this one, David! I recently wrote a similar opinion on my blog - iSCSI has a great niche and FCoE isn't going to kill it.