m5a.16xlarge

Analysis

The m5a.16xlarge is a general purpose instance in the m5a series with 64 vCPUs and 256 GiB of memory on AMD EPYC 7571 (up to 2.5 GHz), which makes it balanced at 4 GiB of memory per vCPU.

Running it in Mumbai instead of São Paulo saves $1,926 a month per instance: $1.78/hour against $4.42/hour, a 148% spread.

It is the cheapest of the 5 processor variants AWS sells at this size and generation. The m5n.16xlarge on Intel Xeon Platinum 8259 (Cascade Lake) costs 38% more, and the m5.16xlarge, the m5ad.16xlarge and the m5d.16xlarge also sit above it.

At $2.75/hour on demand in US East (N. Virginia), about $2,009 a month, the m5a.16xlarge is one of the cheaper ways to buy general purpose capacity: only 14% of the 354 current general purpose types cost less per vCPU-hour than its $0.0430.

Reserving it for a year with no upfront payment cuts the Linux rate to $1.73/hour, which beats on demand once it runs more than 63% of the time; three years brings it to $1.19/hour, 57% below on demand. Windows Server roughly doubles the bill at $5.70/hour, because the licence is charged per vCPU.

It is size 7 of 8 in the m5a series. Stepping up to the m5a.24xlarge gives 96 vCPUs and 384 GiB for $1,004 more a month. Dropping to the m5a.12xlarge saves $502 a month.