I don’t post about technology very often, but here goes. Time to chime in with some industry related info.
There is a new battle in the storage space, FCoE (fibre channel over ethernet) vs. ISCSI (internet SCSI). iSCSI is growing, and is gaining marketshare against FC incumbents, direct attached storage and network attached storage.
To argue that FCoE is superior because of 10GbE and superior delivery mechanisms built around DCE, is dubious at best. It is way too early to say with confidence what the future of FCoE will be. We’re seeing support for 10GbE today with iSCSI so making a GbE iSCSI comparison to 10GbE FCoE is silly. When DCE is resting upon a noisy environment, how efficient will it be? Can it even operate? The point is that we need to compare the technologies in real world environments, from end to end. I personally think, as I’ve indicated earlier, that both technologies will co-exist for some time. Hopefully, the market can be left to decide which solutions solve real problems.
Doug, I have to agree with you, the market should decide which solutions to choose to solve their problems. FCoE solves an important problem: vendors providing FC infrastructure are losing revenues to IP SAN solutions and they need to compensate. 😀
There are a lot of competing ideas, and ways to take advantage of a unified IP infrastructure. There is a place and application for each of these protocols, and hopefully we can come up with a way to make datacenter management easier, whether that means a new protocol, friendlier APIs or standards based management tools. At the end of the day customers are looking for ways to extend their investments, deploy new technologies quickly and well get down to business. If these protocols help, so be it!
I don’t intend to discuss the virtues and applications for iSCSI here, but as it goes in technology formats come and go all the time. In the end I hope the end-users can choose the winner, and not the industry.