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 : IBM -- Ignore unavailable to you. Want to Upgrade?

To: John Koligman who wrote (4183)	10/28/1998 7:08:00 PM
From: Tony Viola	Read Replies (2) \| Respond to of 8218

John, thanks. It's good to see companies building supercomputers out of more standard hardware, like R6/6000 rather than the all custom ones of the old days. Re: "Using one sixth of its total computing power, Blue Mountain was able to run a simulated nuclear weapons test that analyzed physics interactions in an area divided into 30 million zones, he said. A similar simulation on the tried-and-true Cray Y-MP supercomputer was only able to run the simulation with 2.5 million zones." You have to like simulating bombs, rather than exploding them. Also, heaven forbid, maybe we will have more accurate weather predicting some day thanks to computers like this. IBM got some TV ad time publicity over the chess computer; maybe for this one also? Tony

To: John Koligman who wrote (4183)	10/29/1998 9:51:00 AM
From: Arrow Hd.	Read Replies (2) \| Respond to of 8218

Tony, reading through the News.Com article that John posted (which was very good reading) one "architecture/technology" comment I might make is the fact that the SP's "proprietary IBM interconnection hardware" involves, as you might expect from your experience, a lot of microcode (licensed machine code similar to your licensed internal code). It functions similar to the processor controller code on your bi-polar CECs as it controls partitioning MPs into A and B sides or logical partitioning via your MDF feature, or like switch code in a Parallel Sysplex. The Risc code distributes pieces of the job over the available Risc nodes. The more nodes, the faster it gets done. I believe there is some degradation but it is not to the same degree as you see with multi-engine degradation in a commercial mainframe environment.