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 : Silicon Graphics, Inc. (SGI) -- Ignore unavailable to you. Want to Upgrade?


To: Patrick Gainer who wrote (3919)12/30/1997 11:34:00 PM
From: paul  Read Replies (1) | Respond to of 14451
 
So why havent we seen NT scale effectively beyond 2 processors?

I have run into several ERP customers who admit have made a mistake by buying into the NT hype and deployed on PC servers (they will migrate to unix on intel since they cant scrap their investment in intel hardware)- NT 5.0's delays likely into the year 2000 was the final straw since they were basically buying promises. Intel's agreement with Sun around Solaris for Merced is further proof that the industry doesnt believe Microsoft's lies especially when their business is counting on it. Microsoft cant do business as usual when theres viable alternatives for business customers since they dont have a monopoly to squeeze out competion like they did on the desktop.



To: Patrick Gainer who wrote (3919)12/31/1997 10:29:00 AM
From: Alexis Cousein  Read Replies (1) | Respond to of 14451
 
>Things can change overnight.

Building a good scheduler for small, let alone large
SMP or S2MP machines doesn't happen overnight. It took
even Sun years to get where they are now, and I still
think the IRIX scheduler is better.

You are right to say that 'scaling to 16 processors
requires no magic', but doing it so you can actually use the
machines usefully with good throughput, instead of just to
enable bragging, is slightly more complex ;).