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: George Dawson who wrote (25697)1/21/2000 4:56:00 PM
From: Douglas Nordgren  Read Replies (2) of 29386
 
>>Also per previous links here and a paper to be delivered by Ruwart, I believe we have seen that Ancor and Brocade switches can be used in the same SAN.<<

George,

That is indeed the case. The QLGC dual channel HBAs run A & B segmented channels accessing the shared storage. The Brocade and Ancor switches cannot be cascaded to allow A & B cross-connects. Their respective E_Port implementations, though standards compliant, are idiosyncratic. The performances of the A & B channel switches in the LCSE setup are identical and both switches interoperate with all FC components of the SAN. Both can be deployed in the same SAN as long as they don't have to hold hands.

The lack of interop between Ancor and Brocade arises from feature-set implementations to provide product differentiation in the market place. It is important to remember that Ancor and Brocade initially targeted different market segments with their switches. Perhaps a little history lesson is in order.

In the I/O Processor world (FC ASICs are IOPs), the maxim is "software gets you to market faster, you can pit it in the silicon later" (or something like that). Brocade spun an ASIC with Wind River's embedded OS as the command executor, allowing them rapid development and implementation of "options" that populate almost all standards. Their aim was to deliver a robust, fully manageable switch with ease of configuration and installation to OEMs and Channel Vendors who didn't have much in the way of FC expertise or resources to deal with much beyond plug and play. The Silkworms are very much "turnkey" switches.

Ancor's initial target market was the Enterprise Data Center, and they delivered a lean switch which processes I/O fast in Hardware and allowed OEMs to tweak and customize features to provide their own differentiating products. The MkII was very much a "vanilla" switch. Manageability was basic. It was understood that OEMs would want to apply their own management platforms in integrating SANs into their product line.

The issue of the technical superiority of one switch over the the other is moot, as you and Technocrat have stated before. The real differentiation in the companies comes from the application-specific implementation of their products. The ability to manage 1,000 FC switches for 10,000 servers accessing large storage pools is rapidly becoming de rigger (sic).

I am not an engineer, so I don't pretend to know the hardcore details of their respective switch architectures and E_Port characteristics. But this lack of interoperability between switches is not uncommon in the Networking world, as a recent VPN installation taught me.

Things were really jumping this year at Server I/O. Just got back and will post a summary later.

Douglas

PS - The SIO Awards voting was phenomenal, just about 10,000 ballots were received! And the Winners are

Product of the Year 1999

1. CPQ ESS 1200
2. IBM ESS (Shark)
3. EMC Connectrix

Best New Technology for 2000

1. Veritas V3
2. Sun Jiro
3. EMC Control Center

Lifetime Achievement

1. Richie Lary, CPQ
2. Randy Rettberg, Sun
3. Michel Ruettgers, EMC

Thanks to all 10,000 of you who voted to put Ancor in the top five finalists.<G> Note that the best new technologies are all software SAN management products.
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext