m6a.12xlarge

Analysis

The m6a.12xlarge is a general purpose instance in the m6a series with 48 vCPUs and 192 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.12xlarge on AMD EPYC 7571, so the move up a generation is free. The newer m7a.12xlarge on AMD EPYC 9R14 is 34% dearer at $2.78/hour.

Running it in Mumbai instead of São Paulo saves $1,439 a month per instance: $1.33/hour against $3.30/hour, a 148% spread.

Within the same generation and size, the m6g.12xlarge on AWS Graviton2 costs $1.85/hour, 11% less ($165 a month), though switching means rebuilding software for arm64. The other variants are the m6gd.12xlarge at $2.17, the m6i.12xlarge at $2.30 and the m6id.12xlarge at $2.85.

At $2.07/hour on demand in US East (N. Virginia), about $1,514 a month, the m6a.12xlarge 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 $1.37/hour, which beats on demand once it runs more than 66% of the time; three years brings it to $0.9406/hour, 55% below on demand. Windows Server roughly doubles the bill at $4.28/hour, because the licence is charged per vCPU.