Awesome graphics chip by day, the foundation for CUDA by night. NVIDIA viewed GPU computing as a future strength for the company, so G80 led a dual life. The transition to a unified shader architecture gave NVIDIA the chance to, relatively easily, turn G80 into more than just a GPU. Internally, NVIDIA viewed it as an opportunistic way to enable more general purpose computing on its GPUs. I rarely get the opportunity to get a non-marketing answer out of NVIDIA, but in talking to Jonah Alben (VP of GPU Engineering) I had an unusually frank discussion.įrom the outside, G80 looks to be a GPU architected for compute. Much of Fermi's architecture is designed to improve performance in these HPC and other GPU compute applications.Įver since G80, NVIDIA has been on this path to bring GPU computing to reality. When matched with the right algorithms and programming efforts, GPU computing can provide some real speedups. Given these numbers it looks like NVIDIA is building GPUs for a world that doesn't exist. Yet Tesla accounted for less than 1.3% of NVIDIA's total revenue last quarter. The same could be said about GT200, albeit to a lesser degree. There's just one problem:Īs you'll soon see, many of the architectural features of Fermi are targeted specifically for Tesla markets. If NVIDIA met its goals here, the Tesla business could be bigger than the GeForce one. These figures were calculated by looking at the algorithms used in each segment, the number of Hess-like Tesla installations that can be done, and the current budget for non-GPU based computing in those markets. The table below shows what NVIDIA believes is the total available market in the next 18 months for these various HPC segments: Processor Either way the potential is there, it's just a matter of how many similar applications exist in the world.Īccording to NVIDIA, there are many more cases like this in the market. Obviously this doesn't include the servers needed to drive the Teslas, but presumably that's not a significant cost. It migrated a cluster of 2000 servers to 32 Tesla S1070s, bringing total costs down from $8M to $400K, and total power from 1200kW down to 45kW.
One such example is a seismic processing application that HESS found ran very well on NVIDIA GPUs. NVIDIA loves to cite examples of where algorithms ported to GPUs work so much better than CPUs. PCIe connects the S1070 to the host server.
If you have an application that can run well on a GPU, the upside is tremendous.įour of those C1060 cards in a 1U chassis make the Tesla S1070. NVIDIA takes its consumer GPUs, equips them with a ton of memory, and sells them in personal or datacenter supercomputers called Tesla supercomputers or computing clusters. Tesla is NVIDIA's High Performance Computing (HPC) business. Sorry folks, today's launch is targeted entirely at Tesla.Ī GeForce GTX 280 with 4GB of memory is the foundation for the Tesla C1060 cards But that's the extent of what NVIDIA is willing to talk about with regards to Fermi as a gaming GPU. With 3 billion transistors, it had better be.
Data + Geometry Data formats: SEG-Y, SEG-D (including rev.3), SEG-2, 80.Fermi will support DirectX 11 and NVIDIA believes it'll be faster than the Radeon HD 5870 in 3D games.
0 kommentar(er)
