r6a.4xlarge

Analysis

The r6a.4xlarge is a memory optimized instance in the r6a series with 16 vCPUs and 128 GiB of memory on AMD EPYC 7R13 (up to 3.6 GHz), which makes it memory-heavy at 8 GiB of memory per vCPU.

It costs the same as the older r5a.4xlarge on AMD EPYC 7571, so the move up a generation is free. The newer r7a.4xlarge on AMD EPYC 9R14 is 34% dearer at $1.22/hour.

Running it in Mumbai instead of Frankfurt saves $381 a month per instance: $0.5720/hour against $1.09/hour, a 91% spread. It is not offered in São Paulo, so plan around that if you need a presence there.

Within the same generation and size, the r6g.4xlarge on AWS Graviton2 costs $0.8064/hour, 11% less ($73.58 a month), though switching means rebuilding software for arm64. The other variants are the r6gd.4xlarge at $0.9216, the r6i.4xlarge at $1.01 and the r6id.4xlarge at $1.21.

At $0.9072/hour on demand in US East (N. Virginia), about $662 a month, the r6a.4xlarge is one of the cheaper ways to buy memory optimized capacity: only 8% of the 385 current memory optimized types cost less per vCPU-hour than its $0.0567.

It is size 4 of 10 in the r6a series. Stepping up to the r6a.8xlarge gives 32 vCPUs and 256 GiB for $662 more a month. Dropping to the r6a.2xlarge saves $331 a month.