m6a.16xlarge

Analysis

The m6a.16xlarge is a general purpose instance in the m6a series with 64 vCPUs and 256 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.16xlarge on AMD EPYC 7571, so the move up a generation is free. The newer m7a.16xlarge on AMD EPYC 9R14 is 34% dearer at $3.71/hour.

Running it in Mumbai instead of São Paulo saves $1,919 a month per instance: $1.78/hour against $4.41/hour, a 148% spread.

Within the same generation and size, the m6g.16xlarge on AWS Graviton2 costs $2.46/hour, 11% less ($220 a month), though switching means rebuilding software for arm64. The other variants are the m6gd.16xlarge at $2.89, the m6i.16xlarge at $3.07 and the m6id.16xlarge at $3.80.

At $2.76/hour on demand in US East (N. Virginia), about $2,018 a month, the m6a.16xlarge is one of the cheaper ways to buy general purpose capacity: only 17% 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 $1.83/hour, which beats on demand once it runs more than 66% of the time; three years brings it to $1.25/hour, 55% below on demand. Windows Server roughly doubles the bill at $5.71/hour, because the licence is charged per vCPU.