There’s a recent article written comparing the Spec Marks for the Tilera GX series cpus to Intel’s Atom N270 for the first time. Some of these architectures are finally seeing the light of day, running an OS with a test suite to see how they perform. But still Intel is secretly slaving away on it’s answer to Tilera. What will come of this nobody knows because it is even further away from shipping than Tilera GX processors. Reminds me a little of the whole Larrabee time to market problem.
Thus far, Intels Many Integrated Core MIC is little more than a research project. Intel picked up the remnants of the failed “Larrabee” graphics card project and rechristened it Knights and put it solely in the service of the king of computing, the CPU.
Ahhh, alas poor ol’ Larrabee, we hardly knew ye. And yet, somehow your ghost will rise again, and again and again. I remember the hints at the 80 core cpu, which then fell to 64 cores, 40 cores and now just today I read this article to find out it is merely Larrabee and only has a grand total of (hold tight, are you ready for this shocker?) 32 cores. Wait what was that? Did you say 32 cores? Let’s turn back the page to May 15, 2009 where Intel announced the then new Larrabee graphics processing engine with a 32-core processor. That’s right, nothing (well maybe not nothing) has happened in TWO YEARS! Or very little has happened a few die shrinks, and now the upcoming 3D transistors (tri-gate) for the 22nm design revision for Intel Architecture CPUs. It also looks like they may have shuffled around the floor plan/layout of the first gen Larrabee CPU to help speed things up a bit. But, other than these incrementalist appointments the car looks vastly like the model year car from two years ago. Now, what we can also hope has improved since 2009 is the speed and efficiency of the compilers Intel’s engineers have crafted to accompany the release of this re-packaged Larrabee.
I remember the old days when computers ‘re-used’ the system level RAM memory for video on the computer. However the performance disadvantage of doing this was readily apparent when Intel developed video bus technologies VESA Local Bus, AGP, PCIe. So I’m a little surprised a company has developed Flash memory modules for the RAM memory slots on a PC motherboard. What do they hope to gain by building this hybrid disk, memory module. How is seen by the OS? Is it a disk or is RAM memory, many questions.
What a brilliant idea: put flash chips into memory sockets. Thats what Viking Modular is doing this with its SATADIMM product.
This sounds like an interesting evolution of the SSD type of storage. But, I don’t know if there is a big advantage forcing a RAM memory controller to be the bridge to a Flash Memory controller. In terms of bandwidth, the speed seems comparable to a 4x PCIe interface. I’m thinking now of how it might compare to PCIe based SSD from OCZ or Fusion-io. It seems like the advantage is still held by PCIe in terms of total bandwidth and capacity (above 500MB/sec and 2Terabytes total storage). It maybe a slightly lower cost, but the use of Single Level Cell Flash memory chips raises the cost considerably for any given size of storage, and this product from Viking uses the Single Level Cell flash memory. I think if this product ships, it will not compete very well against products like consumer level SSDs, PCIe SSDs, etc. However if they continue to develop the product and evolve it, there might be a niche where it can be performance or price competitive.