We decided to move 90% of our workload from the cloud to on-prem infrastructure

Re:We decided to move 90% of our workload from the cloud to ... by recuter on Tuesday April 12, 2022 @06:26AM (#31000743)

  After spending $100k in a year.. After doing the math, We decided to purchase three workstations at a total cost of $17k. 

  One is a GPU-based workstation with two RTX 3090s and an Intel i9–12900 CPU, and another two workstations with 16 cores AMD Ryzen 5950X CPUs.

  It took us a few FTE days to set those up to our satisfaction with slurm, NFS, backups, and several other services.

  We noticed that our RTXs, although considered gaming cards, are comparable (if not better) in performance to Tesla V100, which some cloud providers rent at the staggering price of $3.06 an hour.

tl;dr - They did the math.

[ Parent | on HN ]

Re:We decided to move 90% of our workload from the cloud to ... by ldoughty on Tuesday April 12, 2022 @06:40AM (#31000823)

They also had a static workload, with no changing requirements, or really any need for scalability.

They also had FTEs with experience to configure the systems (which for a lot of us technical people is a no-brainer, but if you had data scientists with no hardware experience, that might be different)

This worked for them, and I'm happy for them.. the cloud does not solve all problems... It's simply one hardware strategy you can pick.. it's important to review the options!

[ Parent | on HN ]

Re:We decided to move 90% of our workload from the cloud to ... by Spooky23 on Tuesday April 12, 2022 @11:37AM (#31004118)

That’s ok.

If you need a pickup truck two weekends a year, rent. If you need to carry tools and stuff every day, buy!

[ Parent | on HN ]

Re:We decided to move 90% of our workload from the cloud to ... by justsomehnguy on Tuesday April 12, 2022 @06:34AM (#31000791)

> They did the math.

And bought 3 workstations instead of 3 servers.

IMHO the headline of "Why Enzymit Decided to Build its Own On-Prem HPC Infrastructure" is a bit... stretched.

[ Parent | on HN ]

Re:We decided to move 90% of our workload from the cloud to ... by uniqueuid on Tuesday April 12, 2022 @06:55AM (#31000923)

Well, they do use slurm, so it's technically a HPC stack.

[ Parent | on HN ]

Re:We decided to move 90% of our workload from the cloud to ... by raelmebrand on Tuesday April 12, 2022 @06:49AM (#31000880)

This is how universities (or people from academic background) interpret...

[ Parent | on HN ]

Re:We decided to move 90% of our workload from the cloud to ... by HelloNurse on Tuesday April 12, 2022 @09:01AM (#31002058)

They have a workload that is entirely dictated by their protein design etc. projects, not by external access to a web app. So "high performance" means doing a single definite task quickly (so they can *proceed to the next run), not scaling to many users and requests. If they need to scale, they first hire protein designers or the like and then they set up some new workstation, probably something different from existing ones for diversification and obsolescence reasons. No cattle.

[ Parent | on HN ]

Re:We decided to move 90% of our workload from the cloud to ... by dagw on Tuesday April 12, 2022 @08:59AM (#31002029)

We noticed that our RTXs, although considered gaming cards, are comparable (if not better) in performance to Tesla V100

You don't buy Tesla cards because you care about maximizing operations pr second, you buy them because you care about operations pr KWh.

[ Parent | on HN ]

Re:We decided to move 90% of our workload from the cloud to ... by adgjlsfhk1 on Tuesday April 12, 2022 @10:20AM (#31003101)

No, you buy tesla cards so Nvidia doesn't sue you and stop providing firmware.

[ Parent | on HN ]

Re:We decided to move 90% of our workload from the cloud to ... by lostmsu on Tuesday April 12, 2022 @10:24PM (#31010576)

That's a good point. 3090 RTX may be 1.5x faster than my Titan V, but Titan V consumes 2x less energy on the same load while being much older generation card.

[ Parent | on HN ]

Re:We decided to move 90% of our workload from the cloud to ... by dagw on Wednesday April 13, 2022 @03:44AM (#31012127)

Plus less energy consumption means less heat generated which means you can pack them tighter and you can spend less on cooling etc etc. At scale these things really add app.

[ Parent | on HN ]

Re:We decided to move 90% of our workload from the cloud to ... by chrxr on Tuesday April 12, 2022 @07:07AM (#31000988)

Worth noting that using a gaming card for workloads like this would likely void the warranty.

[ Parent | on HN ]

Re:We decided to move 90% of our workload from the cloud to ... by charcircuit on Tuesday April 12, 2022 @07:46AM (#31001284)

No it doesn't

[ Parent | on HN ]

Re:We decided to move 90% of our workload from the cloud to ... by Cpoll on Tuesday April 12, 2022 @05:31PM (#31008460)

Isn't it right there in the fine print?

> Warranted Product is intended for consumer end user purposes only, and is not intended for datacenter use and/or GPU cluster commercial deployments ("Enterprise Use"). Any use of Warranted Product for Enterprise Use shall void this warranty. https://www.nvidia.com/en-us/support/warranty/

[ Parent | on HN ]

Re:We decided to move 90% of our workload from the cloud to ... by charcircuit on Tuesday April 12, 2022 @06:23PM (#31008968)

I must have read over that

[ Parent | on HN ]