News for hackers. Stuff that matters.
Re:Building your own deep learning computer is 10x cheaper t... by soneil on Tuesday September 25, 2018 @05:42AM (#18064799)
I think the analogy holds. The numbers in the article assume constant usage (there is a nod to the cost of static storage, but all the math is done on $3/hr gpu usage, not $0.1/mo/GB storage).

So if your usage is 4 hours per week, 50 weeks a year, you hit $600/year of gpu. If your usage is 24 hours a day, you hit $27k.

So it's not a clear-cut "buying metal is cheaper" - you have to decide how your usage amortizes. As pgeorgi says, if your demand is a pack of cookies a week, buying a bakery isn't the most cost-efficient way to satisfy that demand.

Re:Building your own deep learning computer is 10x cheaper t... by thisisjeffchen on Tuesday September 25, 2018 @09:42AM (#18066216)
Depends on your use case. The more accurate models these days require lots of data with huge models. Machine Translation, Image Generation, Segmentation, etc. Each experiment takes a couple days/weeks to train, so we were encountering long stretches of constant usage.
Re:Building your own deep learning computer is 10x cheaper t... by v_lisivka on Tuesday September 25, 2018 @06:38AM (#18065102)
For DL time is measured in years*GPU instead of hours.
Re:Building your own deep learning computer is 10x cheaper t... by dagw on Tuesday September 25, 2018 @07:49AM (#18065461)
Only for niche problems run into by huge companies and research groups. Most 'normal' problems encountered by 'normal' companies don't need anywhere near that much training.