Intel shows glimpse of 32-core Larrabee beast
Has a larra, larra processing power
Intel has opened up a corner of its kimono and shown a picture of the upcoming Larrabee chip, indicating it will be a 32-core graphics processing engine.
As reported here and elsewhere, Larrabee is Intel's response to Nvidia and AMD graphics processing chips. Larrabee will be, in its first iteration, a 32-core processor. Each core is expected to be an x86 core, and each will will be paired with a vector processing unit. Larrabee will have a shared pool of cache memory and there will be memory interfaces around the edge of the chip.
This is according to people who have inspected the die photo in the Intel slide deck.
At the opening of a Visual Computing Institute at Germany's Saarland University, Joseph Schultz, a VP in Intel's Corporate Technology Group and director of its Microprocessor Technology Lab, said that Larrabee would be launched in the first half of next year, later than the expected late-2009 date.
Intel has made a point of Larrabee's programmability, saying it will be more flexible in use than current dedicated graphics chips.
The idea seems to be that a Larrabee workstation could use the chip for general business applications, mining the humungous x86 application base, as well as power through graphics image rendering tasks. It could also be used for new graphics processing applications because of this flexibility.
New compilers are going to be needed to use Larrabee to its best effect. It is reported that Intel will introduce a packaged set of graphics development software tools called Parallel Studio on May 26th. This is aimed at making the parallelisation of graphics applications easier and is said to help developers using Windows and the Visual Studio product.
Intel said the revealed Larrabee image is just one of what will be multiple versions of Larrabee over time. The firm has published a white paper describing Larrabee here (pdf). This has performance graphs showing up to 64 cores, immediately suggesting Intel could ship versions of Larrabee with differing core counts: 8, 16, 32, 48, and on to 64.
Larrabee is being designed for manufacture using Intel's current 45nm High-K metal gate process, and so we might expect a 32nm version of Larrabee to be on the roadmap, with a consequent possible increase in the core count beyond 64.
If only Intel hadn't already said Larrabee would be using a Pentium CPU core, then we could wonder if Intel is using its Atom processor design as the Larrabee core to meet the chip real estate limitations. That would make Larrabee a molecule made up from Intel atoms. ®
COMMENTS
@Matt
"So whilst Larrabee is a good demonstration of Intel getting really good at multi-core, it's not of much immediate interest."
You'd be an end-user then. This is of *immense* immediate interest to software developers. In the x86 world, it is the most interesting development since 32-bit.
You note that it will run existing apps rather slowly. That's the point. This is the first x86 chip that runs existing software like a dog. (Yeah I remember the Pentium Pro's performance on 16-bit code. This is much worse.) To make Larrabee usable, people need to go back and rewrite that software. How likely is that?
Well, first let me make the obvious point that boxes using this chip will be the first test systems that actually make non-parallel software look bad and parallel software look good. In terms of convincing non-programmers of the need to revisit old code, *both* of those points are important. You'll be able to show your manager, "Look, here's a cheap box with Intel's next-gen chip. The old code runs like molasses in winter but the rewritten code goes faster than the old code did on that super expensive Xeon box over there. So we've got a problem, but the solution could be a huge win."
The last serious temptation to go back and fiddle with existing code was the 64-bit transition. Many folks didn't bother because there was only a marginal performance improvement and there were porting difficulties because the actual semantics of a given line of code might change. This time, Intel are pushing a toolset that (they claim) offers a significant performance boost (on Larrabee, but probably much less so on existing multi-core chips) with *no* semantic changes. That's a free lunch, so we'll treat the claim with some caution, but we do know that at *some* point in the next decade, software is going to experience a revolution. Smart programmers are therefore listening out for the first shot. Maybe this is it.
@Matt
"So whilst Larrabee is a good demonstration of Intel getting really good at multi-core, it's not of much immediate interest."
You can bet is of interest to anyone who ties their software pricing to the number of cores.
I think Intel are talking about several hundred watts for the 32-core offering, but the 8-core Larrabee would be a perfectly reasonable chip to put in an SCC. This could all go mainstream *awfully* fast. Canonical have an OS for such a chip. Microsoft don't, until they change their pricing structure. The same goes for Oracle.
Make no mistake, Larrabee is awfully disruptive technology.
even more nothingness then
Goodie, I'll be able to watch 31 cores doing nothing now, rather than just one.
Ah progress...
Intel shows glimpse of 32-core Larrabee beast
This is for this; http://www.khronos.org/developers/library/overview/opencl_overview.pdf
Art
Pentium P54C core? 32 of them?
What is the name for 32-uple shit:
http://bofh.ntk.net/Star-Trek-Lost.html
