512gb being more expensive than just two 256g is a little strange. You mean you can get the same exact total cluster ram but double the compute/storage nodes and save $700?
What sort of jobs are memory bandwidth bound that people are running? Will they actually see 10x speed improvement or hit some other real world bottleneck? Are these jobs "short" on the order of seconds or minutes or is the node being hammered for two weeks straight? I'm just curious about it.
It might be possible to use that, because you're of course effectively doubling your memory bandwidth at the cost of a slow interlink. (well, still >100 gbit/s, ie. still better than pcie bandwidth in 5090s or RTX6000 card)
[–] asdff 1 day ago
[–] addaon 1 day ago
[–] asdff 1 day ago
https://www.jeffgeerling.com/blog/2025/15-tb-vram-on-mac-stu...
[–] fotta 1 day ago
[–] asdff 1 day ago
[–] addaon 1 day ago
CFD, for me. Depends on how you segment the domain, but the speed hit can be significant.
[–] acchow 1 day ago
[–] speedgoose 1 day ago
[–] spwa4 20 hours ago
[–] HSO 1 day ago