m9g.4xlarge

Analysis

The m9g.4xlarge is a general purpose instance in the m9g series with 16 vCPUs and 64 GiB of memory on AWS Graviton5 (up to 3.3 GHz), which makes it balanced at 4 GiB of memory per vCPU.

Running it in N. Virginia instead of Tokyo saves $167 a month per instance: $0.7827/hour against $1.01/hour, a 29% spread. It is not offered in Canada Central, São Paulo, London and Mumbai, so plan around that if you need a presence there.

It is the cheapest of the 2 processor variants AWS sells at this size and generation. The m9gd.4xlarge on AWS Graviton5 costs 28% more, and no others also sit above it.

On demand it costs $0.7827/hour in US East (N. Virginia), $571 a month if left running, which puts it mid-table for general purpose types at the 32nd percentile per vCPU ($0.0489 per vCPU-hour) and the 31st percentile per GiB ($0.0122 per GiB-hour).

Compared with the previous generation, it costs 9.0% more than the m8g.4xlarge on AWS Graviton4 ($47.19 more a month per instance), in exchange for the newer processor.

Looking beyond its own line, 9 of 37 current types with 16 vCPUs and similar memory are cheaper. The cheapest of those is the m6g.4xlarge on AWS Graviton2 at $0.6160/hour.