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 : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: ted burton who wrote (168878)	7/31/2002 2:14:06 AM
From: wanna_bmw	Read Replies (1) \| Respond to of 186894

Performance Results With EMSL Supercomputer Running Itanium-2 Processors (This organization uses Intel's new Itanium 2 CPU in a supercomputer. It's pretty interesting, and thoroughly impressive. Make sure you give those Itanium guys a high-5, Ted. Also, don't forget to check out this .pdf - performance comparisons vs Power4 inside. emsl.pnl.gov:2080/capabs/mscf/hardware/hpcs2_results.pdf ) Pacific Northwest National Laboratory staff have begun running benchmarks on a .25-teraflop prototype of the newly acquired 9.1-teraflop Hewlett Packard Linux-based supercomputer. This machine is located in the Molecular Science Computing Facility in the William R. Wiley Environmental Molecular Sciences Laboratory, a DOE national user facility. Using a computational chemistry code developed in the EMSL called NWChem that does both electronic structure and molecular dynamics simulations, scientists recorded outstanding performance. Three areas key to computational research -- sustained peak CPU performance, memory bandwidth and interconnectivity -- are showing unparalleled results. In each area, the results were greater than anticipated. The prototype HP supercomputer utilizes Itanium-2 processors, HP's ZX1 chipset and a Quadrics interconnect. "This is the fastest system we've ever run our NWChem codes on, and the results are even more amazing because this is just the prototype," said Scott Studham, operations technical group leader for the Molecular Science Computing Facility at the William R. Wiley Environmental Molecular Sciences Laboratory, a Department of Energy scientific user facility located at PNNL. "This type of performance will take us to the next level in addressing the complex environmental challenges facing the nation." When running a key kernel of NWChem (a matrix multiply), the CPU sustained 96 percent of peak theoretical performance. The prototype HP supercomputer's memory bandwidth was sustained at 3.8 gigabtyes per second for a single CPU and the Quadrics Elan3 interconnect achieved 5 microsecond latency between nodes. The interconnect has demonstrated greater than 320MB per second bandwidth. "These numbers show that our system has powerful memory capabilities, receives data quickly and shares data quickly," Studham said. "That means the scientists who use this system will have an unparalleled ability to address important scientific problems and be able to deliver more reliable predictions. Once the HP supercomputer is completely operational at more than nine teraflops, we expect these key performance metrics to jump even higher than on our prototype." The award-winning NWChem code was developed in the Theory, Modeling and Simulation directorate of the William R. Wiley Environmental Molecular Sciences Laboratory and will be the main code run on the HP supercomputer. EMSL is sponsored by DOE's Office of Biological and Environmental Research. emsl.pnl.gov:2080/capabs/mscf/?/capabs/mscf/hardware/results_hpcs2.html wbmw

To: ted burton who wrote (168878)	7/31/2002 3:52:33 AM
From: Jim McMannis	Respond to of 186894

Ted, RE:"Although I am an Intel employee, I don't speak for my employer" Cool, an Intel employee that follows company policy on bulletin boards...

To: ted burton who wrote (168878)	7/31/2002 8:55:06 AM
From: Dan3	Read Replies (5) \| Respond to of 186894

Re: one part is frail because YOU think it's cheating by using a better thermal solution You've got it backwards. AMD's uncased die is more easily damaged than Intel's cased die, but it also provides a better heat transfer interface. Intel's case protects the die from some installation screw-ups, but can cook the chip during use. If anyone has a better thermal solution, it's AMD, but the trade-off is that an inexperienced installer is a little more likely to damage the chip. Note that AMD used a case on the old K6-2, but elected to go with a bare die for Athlon to improve thermal conduction. Also note that the surface of a sealed box will be at the same temperature as its contents, e.g. the temperature of a die, and the wrapping of a tightly wrapped die, will be the same. It's called the law of conservation of energy....