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 : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: TechieGuy-alt who wrote (27075)	2/1/2001 12:12:38 AM
From: bruce rogers	Read Replies (1) \| Respond to of 275872

>What cache? Do you know that the average linux kernel source runs greater than 15-20 megabytes >Once you include the amount of object files written back to disk (>50 megs!) and the amount of libs read in for the compile Those numbers add up to a lot less than the 256 Megs the guy said the system had.

To: TechieGuy-alt who wrote (27075)	2/1/2001 12:06:12 PM
From: pgerassi	Read Replies (1) \| Respond to of 275872

Dear TG: When a compile takes less than 1/2 the time on a two processor system than a uniprocessor system, the problem is usually traced to cache thrashing. What probably happened is that the working set of the compile process is less than the cache available in one CPU but, when i/o tasks and other OS tasks interrupt the compile process, the working set becomes greater than the cache in the single CPU. In a typical dual system, one CPU spends all of the time handling OS tasks with some left over for compiles. As such it still thrashes the cache less due to the shorter waits between requests (more of them due to two CPUs generating requests) use the critical OS tasks such as disk caching enough to keep it from being evicted from cache. The second CPU spends all of its time in the compile programs and thus, keeps the working set completely within its cache. Thus, it compiles much faster. Both together then do far more than, each one separately. Furthermore, under Linux all memory not used for the OS and running processes is used for caching the disk. This is typically equal to 1/2 to 3/4 of the typical memory configs. My 256MB system, typically has a disk cache of 192MB during kernel compiles. The quickest method to do compiles is not to have X-windows running as it has a large memory footprint. On one client of mine, who has a 1GB Linux box, the cache typically runs over 700MB even, when 12 people are on the system at the same time (some from PCs and some from terminals connected to a multiport serial card (16)). On that system kernel compiles can run completely from the cache (the speed is up to 2x as fast as a system with 1/4th the memory). I hope this clears up your questions on the matter. Pete