7,613
edits
Changes
Jump to navigation
Jump to search
→Hardware
==Hardware==
We mostly followed the original hardware spec from NVIDIA, updating the drives and other minor things, rather trying to work out the boards that go into the more recent variants from Lamdba Labs, Bizon-tech, etc.:
*https://developer.nvidia.com/devbox
*https://lambdalabs.com/deep-learning/workstations/4-gpu
*https://bizon-tech.com/us/bizon-g3000
We had many of these parts available for salvage from other boxes, though we had to buy the ASUS X99-E WS motherboard (as well as some new drives) just for this project. We opted to use a Xeon e5-2620v3 processor, rather than the Core i7-5930K (which we did have available). Both support 40 channels and mount in the LGA 2011-v3 socket, and both have 6 cores, 15mb caches etc. The i7 has a faster clock speed but the Xeon takes registered (buffered), ECC DDR4 RDIMMs, which means we can put 256Gb on the board, rather than just 64Gb. For the GPUs we have a TITAN RTX and an older TITAN Xp available to start, and we can add a 1080Ti later, or buy some additional GPUs if needed. We also put the whole thing in a Rosewill RSV-L4000 case, because we had one available and aren't fussy about how it looks.
{| class="wikitable sortable"
