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 : The New QLogic (ANCR) -- Ignore unavailable to you. Want to Upgrade?

To: KJ. Moy who wrote (21148)	3/18/1999 7:34:00 AM
From: Kelly Igou	Read Replies (2) \| Respond to of 29386

Server clustering using FC as the CPU communication fabric - this would seem to take us one step closer to real-life plug-and-play distributed computing, the way a lot of us conceptualized it in the early '70s. Success will depend more on the systems software's (WolfPack/WolfPup, whatever) ability to manage inter-CPU communications in a low-latency fashion, however, not the speed/latency of the fabric, although the less the fabric contributes to overall latency, the better. In fact, as I think about it, the speed/latency requirements of this application sort of beg for an optical-computing-based FC switch... Note: As the number of processors clustered rises, the communications overhead does not go up arithmetically, it goes up geometrically or exponentially (can't remember which, you'll have to forgive me, it was 24 years ago.) Sounds like a killer on the horizon...!