On: AWS vs. GCP reliability is wildly different

[iLoveOncall]

It is clickbait, the real title should be "AWS vs. GCP on-demand provisioning of GPU resources performance is wildly different".

That said, while I agree that launch time and provisioning error rate are not sufficient to define reliability, they are definitely a part of it.

[tl_donson → iLoveOncall]

“ AWS vs. GCP on-demand provisioning of GPU resources performance is wildly different”

yeah i guess it does make sense that one didn’t win the a/b test

[lelandfe → iLoveOncall]

> wildly different

For this, I'd prefer a title that lets me draw my own conclusions. 84 errors out of 3000 doesn't sound awful to me...? But what do I know – maybe just give me the data:

"1 in 3000 GPUs fail to spawn on AWS. GCP: 84"

"Time to provision GPU with AWS: 11.4s. GCP: 42.6s"

"GCP >4x avg. time to provision GPU than AWS"

"Provisioning on GCP both slower and more error-prone than AWS"

[esrauch → lelandfe]

84 of 3000 failed is only "one nine"