RE:
itjungle.com
Edited for emphasis:
1. Brandt says that the way that the Opterons will be linked to the Cells has not been precisely determined
2. A forthcoming Cell-based blade server, which IBM was prototyping earlier this year, is expected to be launched for IBM's BladeCenter chassis in a matter of weeks, and these will be used in the Roadrunner machine. The Opteron portion of the supercomputer will be comprised of System x3755 servers, which Big Blue announced in early July. These are four-socket boxes, which can therefore support up to eight Opteron cores using the new Rev F Opteron processors, which were announced in late July.
3. Los Alamos plans to build Roadrunner in three stages.
Stage 1: Opteron machines installed, 16,000 Opteron cores = 80 TFlop
Stage 2 early 2007: add some Cell-base blade servers and lots of programming assistance to tune Linux [Red Hat Fedora] for the Cell chip ... architecting the machine and tuning it. Work will be done to make compilers automatically vectorize code for the Cell chips.
Stage 3: 8,000 Cell-based blade servers, with two Cell chips on each blade, will be added to Roadrunner, adding the bulk of the 1.6 peak petaflops of computing power it comprises.
Brandt said that IBM was still mulling the way the Opteron and Cell machines will connect to each other. IBM could have added an HTX HyperTransport port to the Cell blade and plugged that into the LS42 blade, which uses HTX ports to link two two-socket blades together to make a four-socket SMP blade. But, for whatever reason, IBM decided to not do this.
The Cell blade uses an IBM-developed bus architecture called the element interconnect bus, or EIB, to link the Cell vector processors to each other and to link Cell chips to each other into processor fabrics, and this interconnect may be incompatible with HyperTransport links, which are used by AMD to link multiple Opterons together. Brandt says that by choosing the four-socket rack-mounted server instead of Opteron-based blades, Los Alamos and IBM can keep their connectivity options between the two types of machines open.
--------
My read is that CELL will be the "dominant" chip in HPC if IBM pulls of the compiler auto-vectorization for CELL. Opteron becomes a manager of the CELL hive or executer of the portion of processes that need out of order execution.
IBM will use only as many Opterons needed to make the CELL blades shine. A symbiosis, perhaps, but not a marrage of equals.
-Magrathea |
