r8a.12xlarge

Analysis

The r8a.12xlarge is a memory optimized instance in the r8a series with 48 vCPUs and 384 GiB of memory on AMD EPYC 9R45 (up to Up to 4.5 GHz), which makes it memory-heavy at 8 GiB of memory per vCPU.

Within the same generation and size, the r8g.12xlarge on AWS Graviton4 costs $2.83/hour, 26% less ($735 a month), though switching means rebuilding software for arm64. The other variants are the r8i-flex.12xlarge at $3.17, the r8i.12xlarge at $3.33 and the r8gd.12xlarge at $3.53.

Running it in N. Virginia instead of Frankfurt saves $577 a month per instance: $3.83/hour against $4.63/hour, a 21% spread. It is not offered in São Paulo, London, Mumbai, Singapore and Sydney, so plan around that if you need a presence there.

On demand it costs $3.83/hour in US East (N. Virginia), $2,799 a month if left running, which puts it mid-table for memory optimized types at the 48th percentile per vCPU ($0.0799 per vCPU-hour) and the 64th percentile per GiB ($0.00999 per GiB-hour).

Looking beyond its own line, 20 of 40 current types with 48 vCPUs and similar memory are cheaper. The cheapest of those is the r6g.12xlarge on AWS Graviton2 at $2.42/hour.

Compared with the previous generation, it costs 5.0% more than the r7a.12xlarge on AMD EPYC 9R14 ($133 more a month per instance), in exchange for the newer processor.