Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Linode GPU Instances

blog.linode.com

Post reply on HN
Print view View original

1–10 of 75 posts

Re: Linode GPU Instances

#3

Post by ksec »

minimaxir wrote: post #2

Huh. Given that cheap cloud GPUs are nowadays sought for training AI, launching with a Workstation-oriented GPU is an odd product decision.

Are there any difference? Seems to support CUDA as well, I don't see anything wrong with it.

Also seems to be a lot cheaper than AWS counterpart.

Re: Linode GPU Instances

#4

Post by minimaxir »

ksec wrote: post #3
minimaxir wrote: post #2

Huh. Given that cheap cloud GPUs are nowadays sought for training AI, launching with a Workstation-oriented GPU is an odd product decision.

Are there any difference? Seems to support CUDA as well, I don't see anything wrong with it. Also seems to be a lot cheaper than AWS counterpart.

It seems like a Quadro RTX 6000 is comparable performance-wise to a P100 (https://askgeek.io/en/gpus/vs/NVIDIA_Quadro-RTX-6000-vs-NVID...), although there are no tests for the ML angle.

A P100 on GCP is $1.46/hr alone, so maybe Linode is a good deal if the performance is indeed comparable.

Re: Linode GPU Instances

#7

Post by keytarsolo »

That pricing isn't too bad. They come with decent SSD storage too, which is key for the large datasets that make a GPU instance worthwhile.

Linode skews more towards smaller scale customers with many of their offerings so I think the GPUs here make sense. The real test will be how often they upgrade them and what they upgrade them too.

Re: Linode GPU Instances

#8

Post by thenightcrawler »

Isn't AWS cheaper?

edit: could be wrong thought I read of AWS being .65 dollars an hour for deep learning GPU use. edit2: Did a quick look, the .65 dollars doesn't include the actual instance, so its around 1.8 an hour on the low end, I think this cheaper.

Post reply on HN