4 ms·

what's the per-hour cost spot price of this machine on AWS ?

by nqzero 6y ago

ChuckMcM 6y ago
well an m5ad.24xlarge is 96 threads and 384G with your own 2xSSD (saves on EBS bandwidth costs). So fewer threads but a bit more memory. (we'll guess that is a 48 core EPYC 7642 equivalent with 96 threads since there is no 96 core version) That bad boy costs $0.96/hr on the current spot prices (https://aws.amazon.com/ec2/spot/pricing/ https://aws.amazon.com/ec2/spot/pricing/)
- dmurray 6y ago
  How does having your own SSD save on bandwidth? You have to send the data to and from the cores where the computation is taking place. There are certainly some workloads where it makes sense to own your own storage and rent computation, but you can't assume that by default for a "powerful AI" workload.
  - cavisne 6y ago
    It saves on bandwidth $ costs to EBS (AWS block storage)
- qayxc 6y ago
  That instance type is missing the RTX 2080 Ti-equivalent GPU, though... The closest in performance to that GPU would be the V100 found in P3-instances.
  - sabalaba 6y ago
    Yea which are $3 / hour. Or, you could, shameless plug, use Lambda's GPU cloud which has V100s for $1.5 / hour! https://lambdalabs.com/service/gpu-cloud https://lambdalabs.com/service/gpu-cloud I'm embarrassed that I feel so compelled to post this--on a Friday night at that--I apologize.
    - manjunaths 6y ago
      > I'm embarrassed that I feel so compelled to post this--on a Friday night at that--I apologize. Don't be. Finding gems like this is why some of us read the comments.
      - 6510 6y ago
        Frowning on self-promotion got us in those silos.
qayxc 6y ago
There is no exact cloud equivalent - the researchers used commodity hardware for their GPU, something that NVIDIA doesn't allow for use in data centres. The closest you can get on AWS (more like System #3 in the paper with 4x GPUs) would be something like a p3.8xlarge instance [1] that'll cost you $12.24 (on demand) or $3.65 to $5 (spot price, region dependent) [2]. A single GPU instance (p3.2xlarge) only 16 vCores, though) will cost you $3.06 on-demand or $0.90 to $1.20 (spot). [1] https://aws.amazon.com/ec2/instance-types/p3/ https://aws.amazon.com/ec2/instance-types/p3/ [2] https://www.aws4less.com/AWS/AWSCurrentSpotPrice.php https://www.aws4less.com/AWS/AWSCurrentSpotPrice.php
rasz 6y ago
around 3-4 months on AWS would equal buying hardware on the spot.
molszanski 6y ago
The cost of a similarly spec’d machine on AWS is 0.73 kidney/hr

what's the per-hour cost spot price of this machine on AWS ?

by nqzero 6y ago

what's the per-hour cost spot price of this machine on AWS ?

well an m5ad.24xlarge is 96 threads and 384G with your own 2xSSD (saves on EBS bandwidth costs). So fewer threads but a bit more memory. (we'll guess that is a 48 core EPYC 7642 equivalent with 96 threads since there is no 96 core version) That bad boy costs $0.96/hr on the current spot prices (https://aws.amazon.com/ec2/spot/pricing/ https://aws.amazon.com/ec2/spot/pricing/)

dmurray 6y ago
How does having your own SSD save on bandwidth? You have to send the data to and from the cores where the computation is taking place. There are certainly some workloads where it makes sense to own your own storage and rent computation, but you can't assume that by default for a "powerful AI" workload.
- cavisne 6y ago
  It saves on bandwidth $ costs to EBS (AWS block storage)
qayxc 6y ago
That instance type is missing the RTX 2080 Ti-equivalent GPU, though... The closest in performance to that GPU would be the V100 found in P3-instances.
- sabalaba 6y ago
  Yea which are $3 / hour. Or, you could, shameless plug, use Lambda's GPU cloud which has V100s for $1.5 / hour! https://lambdalabs.com/service/gpu-cloud https://lambdalabs.com/service/gpu-cloud I'm embarrassed that I feel so compelled to post this--on a Friday night at that--I apologize.
  - manjunaths 6y ago
    > I'm embarrassed that I feel so compelled to post this--on a Friday night at that--I apologize. Don't be. Finding gems like this is why some of us read the comments.
    - 6510 6y ago
      Frowning on self-promotion got us in those silos.

qayxc 6y ago

There is no exact cloud equivalent - the researchers used commodity hardware for their GPU, something that NVIDIA doesn't allow for use in data centres. The closest you can get on AWS (more like System #3 in the paper with 4x GPUs) would be something like a p3.8xlarge instance [1] that'll cost you $12.24 (on demand) or $3.65 to $5 (spot price, region dependent) [2]. A single GPU instance (p3.2xlarge) only 16 vCores, though) will cost you $3.06 on-demand or $0.90 to $1.20 (spot). [1] https://aws.amazon.com/ec2/instance-types/p3/ https://aws.amazon.com/ec2/instance-types/p3/ [2] https://www.aws4less.com/AWS/AWSCurrentSpotPrice.php https://www.aws4less.com/AWS/AWSCurrentSpotPrice.php

rasz 6y ago

around 3-4 months on AWS would equal buying hardware on the spot.

molszanski 6y ago

The cost of a similarly spec’d machine on AWS is 0.73 kidney/hr