Author: Josh Patterson
Date: 2/24/2021
This is a calculator to give you a rough estimate for what your team's GPU costs on AWS for Deep Learning based on their usage patterns. Obviously there can be more factors involved, but this is meant to give you a ballpark figure to start a conversation from to estimate costs.
See our blog post on "Forecasting Your AWS GPU Cloud Spend for Deep Learning" for more detailed discussion around how we make our estimations and calculations.
Team Label | Data Scientist (DS) Count | GPU Hours per Month per DS | GPU Instance Type | SageMaker Multiple | GPU Hours per Year | Group AWS Subtotal Cost per Year | |
---|---|---|---|---|---|---|---|
1. | 0 | $0.00 | |||||
2. | 0 | $0.00 | |||||
3. | 0 | $0.00 |
Would you like a customized cloud GPU spend analysis for your organization? Reach out and we’ll provide your company this GPU usage analysis report free of charge.
In the table below we show AWS' pricing for the p3 instances.
Instance Size | GPUs - Tesla V100 |
GPU Peer to Peer |
GPU Memory (GB) | vCPUs | Memory (GB) | Network Bandwidth | EBS Bandwidth | On-Demand Price/hr* | 1-yr Reserved Instance Effective Hourly* | 3-yr Reserved Instance Effective Hourly* | Blended Price Avg(OnDemand and 1yr Reserved) |
---|---|---|---|---|---|---|---|---|---|---|---|
p3.2xlarge | 1 | N/A | 16 | 8 | 61 | Up to 10 Gbps | 1.5 Gbps | $3.06 | $1.99 | $1.05 | $2.53 |
p3.8xlarge | 4 | NVLink | 64 | 32 | 244 | 10 Gbps | 7 Gbps | $12.24 | $7.96 | $4.19 | $10.10 |
p3.16xlarge | 8 | NVLink | 128 | 64 | 488 | 25 Gbps | 14 Gbps | $24.48 | $15.91 | $8.39 | $20.20 |
p3dn.24xlarge | 8 | NVLink | 256 | 96 | 768 | 100 Gbps | 19 Gbps | $31.22 | $18.30 | $9.64 | $24.76 |
We've added a column at the end where we've averaged the price of On-Demand instance pricing and 1-Year Reserved Instances. For this cost model this is the hourly cost we use for instances because we want to account for the fact that some AWS resources will be used by good forecasting and 1-Year Reserved instances at the lower price. However, we also want to account for users who will end up using On-Demand instances, so we average the two prices together.