m6a.24xlarge

Analysis

The m6a.24xlarge is a general purpose instance in the m6a series with 96 vCPUs and 384 GiB of memory on AMD EPYC 7R13 (up to 3.6 GHz), which makes it balanced at 4 GiB of memory per vCPU.

It costs the same as the older m5a.24xlarge on AMD EPYC 7571, so the move up a generation is free. The newer m7a.24xlarge on AMD EPYC 9R14 is 34% dearer at $5.56/hour.

Running it in Mumbai instead of São Paulo saves $2,879 a month per instance: $2.67/hour against $6.61/hour, a 148% spread.

It is the cheapest of the 5 processor variants AWS sells at this size and generation. The m6idn.24xlarge on Intel Xeon 8375C (Ice Lake) costs 84% more, and the m6i.24xlarge, the m6id.24xlarge and the m6in.24xlarge also sit above it.

At $4.15/hour on demand in US East (N. Virginia), about $3,027 a month, the m6a.24xlarge is one of the cheaper ways to buy general purpose capacity: only 16% of the 354 current general purpose types cost less per vCPU-hour than its $0.0432.

Reserving it for a year with no upfront payment cuts the Linux rate to $2.74/hour, which beats on demand once it runs more than 66% of the time; three years brings it to $1.88/hour, 55% below on demand. Windows Server roughly doubles the bill at $8.56/hour, because the licence is charged per vCPU.