c7gn.8xlarge

Analysis

The c7gn.8xlarge is a compute optimized instance in the c7gn series with 32 vCPUs and 64 GiB of memory on AWS Graviton3 (up to 2.6 GHz), which makes it compute-heavy at 2 GiB of memory per vCPU.

Within the same generation and size, the c7g.8xlarge on AWS Graviton3 costs $1.16/hour, 42% less ($611 a month). The other variants are the c7i-flex.8xlarge at $1.36, the c7i.8xlarge at $1.43 and the c7gd.8xlarge at $1.45.

Compared with the previous generation, it costs 44% more than the c6gn.8xlarge on AWS Graviton2 ($449 more a month per instance), in exchange for the newer processor. The newer c8gn.8xlarge on AWS Graviton4 is 5.0% cheaper at $1.90/hour, which makes the c7gn.8xlarge hard to justify for new deployments.

Running it in N. Virginia instead of Tokyo saves $381 a month per instance: $2.00/hour against $2.52/hour, a 26% spread. It is not offered in Canada Central, São Paulo, Frankfurt, Mumbai, Singapore and Sydney, so plan around that if you need a presence there.

Per vCPU it is one of the dearer compute optimized types: 87% of the 291 current types in the family cost less than its $0.0624 per vCPU-hour. The premium mostly pays for the very fast networking. On demand it is $2.00/hour in US East (N. Virginia), or $1,458 for a 730-hour month.