g5g.8xlarge

Analysis

The g5g.8xlarge is a gpu instance instance in the g5g series with 32 vCPUs and 64 GiB of memory on AWS Graviton2 (up to 2.5 GHz), which makes it compute-heavy at 2 GiB of memory per vCPU.

Running it in N. Virginia instead of Singapore saves $400 a month per instance: $1.37/hour against $1.92/hour, a 40% spread. It is not offered in Ohio, Canada Central, São Paulo, Ireland, London, Mumbai and Sydney, so plan around that if you need a presence there.

It carries 1 GPU with 16 GB of GPU memory, so most of the price is the accelerator rather than the CPUs. Storage is EBS only. Network: Up to 12 Gigabit.

At $1.37/hour on demand in US East (N. Virginia), about $1,002 a month, the g5g.8xlarge is one of the cheaper ways to buy gpu instance capacity: only 1% of the 78 current gpu instance types cost less per vCPU-hour than its $0.0429.

It is the cheapest of the 2 processor variants AWS sells at this size and generation. The g5.8xlarge on AMD EPYC 7R32 costs 78% more, and no others also sit above it.

It is size 4 of 5 in the g5g series. Stepping up to the g5g.16xlarge gives 64 vCPUs and 128 GiB for $1,002 more a month. Dropping to the g5g.4xlarge saves $397 a month. Unit price is not flat across the series ($0.0429 to $0.1050 per vCPU-hour), so the size changes the rate you pay.