Silicon Investor (SI) -- The First Internet Community

STOCKTALK

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 : Son of SAN - Storage Networking Technologies -- Ignore unavailable to you. Want to Upgrade?

To: Srinivasan Balasubramanian who wrote (1554)	10/31/1999 1:14:00 AM
From: Douglas Nordgren	Read Replies (2) \| Respond to of 4808

>>This switch imo has terrific potential as the switching backplane is separated from the protocol engine and the ports that come in as a PIM. In other words, based on the same architecture, Capellix in the future can double up as a backbone fabric switch. Of course, I could be wrong but I would like to know where I am wrong.<< Srini, I believe you are correct as regards to the future extensibility of the Blades (future PIMs might allow ATM/FC-AL and/or FC-SW/FC-AL interconnects). WRT its current implementation, I regard the Cappellix as a FC-AL Switch (a Super-Loop Director?) which can directly replace a FC-AL hub or consolidate several FC-AL hubs within the backplane. There is definitely a market for a hub-killer. The White paper emphasises the complexity and cost of integrating FC-AL plants into a FC-SW fabric and the Capellix is meant to obviate the headaches of swapping out controllers, drivers, software, configuration, interoperability etc. This might have been true back in March when the paper was authored. Come next week, with expected product announcements from Ancor and Brocade, the next iteration of FC switches may be engineered to address these concerns. We may see then if the per-port-cost and complexity arguments will still hold water. I am no engineer, but I note 2 items in the Capellix tech specs that caught my eye, latency and wavelength. Port-to-port latency could be a problem in larger installations, and shortwave transceivers allow a max of "only" 500 meters between nodes. 5 football fields may cover distributed storage on a campus with minimal latency effect. For extensibility beyond that, I believe Capellix' FC-SW (fabric) Blade will be required to interoperate with other fabric switches (the backplane will pass the loops to the fabric blade which in turn will link to a fabric switch). As always, I stand to be corrected by anyone who knows better. Regards and good luck with ZOOX. This FC sea change floats a lot of boats. Douglas