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.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: Daniel Schuh who wrote (120912)	7/25/2000 5:19:32 PM
From: pgerassi	Respond to of 1572159

Dear Dan: Re: 64 bit addressing This is not true in Solaris x86. Each process is not started at 0. System calls get their own stack per call (limited to 4K until V8.0 (I ran into to this problem recently)). Their reason was that the Physical Memory of any x86 machine is limited to 4GB. Virtually, in 386 and better, the limit is 64TB (including K7). In those servers where more memory is available, NUMA (Non Uniform Memory Access) is used (a portion say 2GB is given to each CPU and 2 GB is shared among them all is one example (this is how it was done before x86 with Z80s called Banking)). The number of processes was allocated statically during kernel compile/link. But for very large servers, 64 bits allows a HUGE virtual and physical space. 16 Quintabytes (2^64) will not be available except virtually for a long time even with todays rapid rise in physical memory sizes. In the old days, 1MB could run 16 users easy having 256 processes. Thus 4GB of physical memory could hold millions of processes no problem except they ran slow (with ASR-33 TTYs who knew?). Pete