Brocade says if FCoE doesn't happen, it will be ready
Financial analysts get to hear interesting snippets
Hold the front page: Brocade told financial analysts that the Fibre Channel over Ethernet (FCOE) transition may not happen. Oh, and it said it had messed up an IBM relationship.
The things financial analysts get told. These revelations came from Brocade's chief marking officer John McHugh who was telling it how it is at a Stifel Nicolaus Technology Conference yesterday.
He discussed what Stifel's report calls "continued lacklustre FCoE adoption." FCoE is the running of Fibre Channel storage networking block-access protocol over Ethernet instead of using physical Fibre Channel cabling and switchgear. It has been, is being, assumed that this transition to Ethernet would happen, admittedly taking several years, because Ethernet is cheap, steamrolls all networking opposition, and is being upgraded to provide the reliable speed and lossless transmission required by Fibre Channel-using devices.
But hold on to your Ethernet horses: this may not be the case. McHugh's FCoE remarks were represented like this:
In terms of FCoE, while not noting that this transition may not ultimately take place, Mr McHugh noted that it is important for the company to invest in this potential transition in the case it were to materialise going forward.
He may not have actually said that the FCoE transition may not take place but by saying Brocade should invest in this "potential transition ... if it were to materialise" he is effectively saying the same thing.
FCoE could be a bust, a waste of time. Storage Fibre Channel protocol messages need to be on a separate network, even with FCoE, because you can't have your storage networking traffic screwed by disruptions in the underlying Ethernet wiring and switch infrastructure. If you have to invest in that then why, as an existing Fibre Channel user, should you bother?
Stifel's report said that Brocade: "currently sells primarily 8Gbps solutions with the company having roughly a 66 per cent share in 8Gbit/s with approximately 50 per cent of total customers actually using 8Gbit/s solutions."
McHugh said a doubling of Fibre Channel speed was coming – from 8Gbit/s to 16Gbit/s. Brocade reckons it has a 12 to 18 month lead over the competition with this technology upgrade, but didn't provide a delivery roadmap. What it means is that there is no need for Fibre Channel customers to go to Ethernet for speed; in fact, the reverse is true. Stick with physical Fibre Channel and have your storage access run faster to keep up with flash-enhanced storage arrays and virtualised, multi-core servers.
IBM and Dell
McHugh dropped another little surprise... Brocade had messed up with IBM, failing to achieve all its goals for that relationship. He said that Brocade was intent on using this failure as a learning experience to boost its Dell relationship; Dell both OEMs and resells Brocade equipment. He said that business with Dell was good and was incremental. It would not have come in without the Dell deal, he added.
IBM bought Ethernet equipment supplier Blade Network Technologies in September 2010.
Networking supplier Juniper also has a Dell relationship but, in Dell design win terms when compared to Juniper, Brocade thinks it appears to be performing well. ®
Thats what youy get when trying to combine a more or less comparable technology with a totally different strategic, tactical and operational culture. It will not work.
I have nothing against the technology itself, I think it's great but not much thought has been put into the management side of things.