3 ms·
Intel tried that with Intel Xeon Phi, basically.
by znpy 6y ago
Intel tried that with Intel Xeon Phi, basically.
- fortran77 6y agoThat was just a lot of Intel cores. I'm talking about an NVIDIA GPU that happens to have some ARM cores along with the CUDA cores.
- kkielhofner 6y agoYou're essentially talking about a Jetson: https://elinux.org/Jetson_AGX_Xavier https://elinux.org/Jetson_AGX_Xavier Eight Nvidia ARMv8 cores, 512 Volta cores, 64 tensor cores, and 2 DLAs - all connected to 32 GB of shared RAM. It was announced over two years ago and the price has fallen to $800. For another $200 you can add a USB-C to DP cable (for a second display) and 1TB of NVMe storage. If I had a $1,000 compute budget it's what I'd get. You'd have a more than functional dual 4K daily driver for dev work, browsing, etc plus the ability to run ML workloads. Oh yeah, and it tops out at less than 50 watts from the wall with all but the heaviest tasks coming in at 25-50% of that.
- fortran77 6y agoI love the Jetson -- we have a bunch of them here -- but that's a SBC. What I want is an NVIDIA 2080Ti or 3080Ti card with some ARM cores that I can run apps that use CUDA+ARM directly on!
- kkielhofner 6y agoI love my Jetsons too! One has to wonder if at some point Nvidia will produce a PCIe carrier board or something for souped up Jetson modules or some variant. I have no idea how the connectivity (PCIe device?) would work out but it's definitely within the realm of possibility.
- my123 6y agoThe Xavier generation supports behaving as a PCIe device (PCIe device mode), so it's maybe not too outlandish. :-)