SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : The New QLogic (ANCR)
QLGC 16.070.0%Aug 24 5:00 PM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: J Fieb who wrote (23270)7/11/1999 10:14:00 AM
From: KJ. Moy  Read Replies (2) of 29386
 
J Fieb,

<<The architecture of this 21st century datacenter
will consist of a central Storage Area Network (SAN) woven together with optical fiber and connected to the intranet/extranet/internet and attended to by a halo of clustered processing nodes consisting of between dozens and thousands of 64-bit enterprise-class microprocessors. Management will be completely data-centric and network-centric and processor vendors will compete for attachment rights to this SAN. >>

In order to fully take advantage of this scenario, new-world-order, applications have to written and used in a special way. Think about the following scenario.

Many processors and many storage devices are clustered together in a SAN. Processors and storage devices are added as needed. However, the possibility of failed components are greater now. So, in a normal operation, an application would run on all processors with a share loading environment. When a processor goes down, all other processors will share the additional load without interrupting operation. Some applications today are already written that way. Only an alert would be sent to a SAN management device for proper action. Storage devices are a little different. When a cluster of storage devices are down, the data in those devices are unavailable. The question is, how many duplicate copies of these devices or just important files need to be stored somewhere else in the same/different SAN? The planning for this scenario is no small task. Every case could be somewhat different depending on how much a datacenter want to spend. And, the expertise and planning initiative generally reside with the processor(i.e. application) makers.

What does this all mean to FC? It is going to be phenomenal. Kurt is right. As the cost of FC switches come down, why would anyone go to FC hubs? Both Brocade and Ancor will prosper.

KJ
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext