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 : Rambus (RMBS) - Eagle or Penguin -- Ignore unavailable to you. Want to Upgrade?

To: Steve Lee who wrote (44103)	6/11/2000 10:36:00 PM
From: Scumbria	Read Replies (1) \| Respond to of 93625

Steve, You are wrong. Please stop distorting facts. Your comments about "effective memory bandwidth" including pagefile considerations are complete waffle. It's not a case of paging to disk, it's a case of whether data has to be read from disk, or from a file/database cache in RAM. The important RAM consideration in a server is most definitely capacity rather than bandwidth. If a page of a file is held in DRAM, it is paged in. If it is on disk, it is paged out. I realize that you are having a little trouble grasping the concept, but the reason to have a lot of fast DRAM is to maximize bandwidth. Please stop accusing me of distortion. That is not the problem here. Scumbria

To: Steve Lee who wrote (44103)	6/12/2000 1:13:00 AM
From: jim kelley	Respond to of 93625

Good points Steve, The disk subsystem is so much slower(10- 20X) than the main memory that the more of the working set you can hold in memory the better the performance. The memory bandwidth issue affects the processor memory interface more than the storage memory system interface. If the working set is in main memory then processing can benefit from higher bandwidth memory. Given two choices first go for a lot of memory and second go for high bandwidth memory for best results.