Ask HN: Cheapest option to colo a GPU compute box?

6 pointsSeckinJohn11y ago5 comments

Hi everyone, We need to build a GPU compute box (thinking GeForce GTX Titan Black[$1000] as the GPU) for training Convolutional Neural Networks and building services on top of the trained model. What would be the most cost effective way to colo such a server? Hurricane Electric told us a 2U rack space (100mbit and 750W) would cost us $700/month and I find that a little too expensive.(they are saying that we would rent a whole cabinet for the same price since their main cost is going to be our power usage and not bandwidth etc)

I am also thinking of building the box on my own and putting it in our office in SF (which has a static ip and 50mbit unlimited internet) but that'd probably violate the TOS - and even if it doesn't I am not sure if the internet connection would be stable enough to operate such a service from (though it doesn't necessarily need to be real-real-time -- we just the box to consume our job queue and a 1 minute delay would be acceptable there)

What are your thoughts on this? Do we really have to come to terms with having to pay $700+/mo for this?

(somehow arranging this for ~$300/mo would be amazing)

5 comments

mattbillenstein11y ago

I'd stick it in your office - $300/mo is still $3600/yr...

Other option I'd look at is spot gpu instances on aws - then it's hosted and you don't buy anything - and potentially you could shut it down part of the time although you're charged a full hour for each partial hour...

mithras11y ago

I don't know about AWS for this use case, if you use it more than 50% of the time I'm sure costs will be much much higher.

SeckinJohnOP11y ago

Actually he is right - gpu instances are pretty cheap if you use the spot ones. http://ec2price.com/?product=Linux/UNIX&type=g2.2xlarge&regi...

But the thing is I am not sure if amazon keeps a spot-instance reserve (i.e. what happens if all the spot instances get turned into on-demand ones over, say, a month and suddenly your service is shut down and you keep waiting indefinitely[you'd probably wait for around 24 hours and then try to get an on-demand one - which is 11x more expensive, costing around $600/mo for a shared gpu. and if you can't find an on-demand one, you will be offline for around a week until you figure out a way of provisioning a gpu instance from another service provider, which will probably cost you a lot more, but at that point you will have to do it not to disrupt the service for so long])

walterbell11y ago

How about sonic.net business class symmetric DSL which permits servers, http://www.businessweek.com/articles/2013-09-26/sonic-dot-ne...

SeckinJohnOP11y ago

Thanks, this looks great. Apparently they also have colocation space -- asked them how much it'd cost us. Will post here once I hear back(if I hear back soon :) ).

j / k navigate · click thread line to collapse

Ask HN: Cheapest option to colo a GPU compute box?

6 pointsSeckinJohn11y ago5 comments

What are your thoughts on this? Do we really have to come to terms with having to pay $700+/mo for this?

(somehow arranging this for ~$300/mo would be amazing)

5 comments

mattbillenstein11y ago

I'd stick it in your office - $300/mo is still $3600/yr...

mithras11y ago

I don't know about AWS for this use case, if you use it more than 50% of the time I'm sure costs will be much much higher.

SeckinJohnOP11y ago

Actually he is right - gpu instances are pretty cheap if you use the spot ones. http://ec2price.com/?product=Linux/UNIX&type=g2.2xlarge&regi...

walterbell11y ago

How about sonic.net business class symmetric DSL which permits servers, http://www.businessweek.com/articles/2013-09-26/sonic-dot-ne...

SeckinJohnOP11y ago

Thanks, this looks great. Apparently they also have colocation space -- asked them how much it'd cost us. Will post here once I hear back(if I hear back soon :) ).

j / k navigate · click thread line to collapse