2 ms·
> AWS built an incredibly rare dedicated compute cluster completed in August 2022 to the requirements of Stability. The 4,000 A100 cluster has dedicated capacit
by orf 3y ago
> AWS built an incredibly rare dedicated compute cluster completed in August 2022 to the requirements of Stability. The 4,000 A100 cluster has dedicated capacity on a single spline which optimizes performance and is one of the only ones there
This wording really doesn’t help your case. It sounds like a total exaggeration.
Link to the AWS case study that highlights this relationship, or absolutely anything than just repeating “we are one of the only companies to have AWS buy a million GPUs for us and let us use them and it’s one of the only ones in this part of AWS’s network. Source: trust me bro”
- emadm 3y agoWe actually have a blog post coming on specifically this from AWS, at the time it was the 8th largest public A100 cluster and we covered it live again while we were preseed well before the funding round https://twitter.com/emostaque/status/1553120829336129537?s=46 https://twitter.com/emostaque/status/1553120829336129537?s=4... It was super interesting as an experience. At the very least anyone who has used AWS for ML can tell you that getting a 40 A100 quota let alone 4,000 is a feat in itself ^_^