The c8g.48xlarge pairs 192 vCPU with 384 GiB of RAM, the 2.00 GiB-per-vCPU ratio that marks a compute-optimized instance. In the name, "c" stands for compute, "8" is the generation, "g" marks AWS Graviton (ARM) processors and "48xlarge" is the size within the c8g family. It runs on AWS Graviton4 Processor at 2.7 GHz, and AWS introduced it in September 2024. AWS has since shipped c9g; c9g.48xlarge is the direct replacement at this size, scoring 6,868,181 CoreMark/s against this instance's 5,350,425 — roughly 28% more throughput for the same vCPU count. Spot's node ladder tops out below this size, so c8g.48xlarge becomes 12 × ch.vs1.2xlarge (192 vCPU, 384 GiB combined), about 94% below the US East (N. Virginia) on-demand rate. Spot delivers compute as managed Kubernetes, so a node pool of that size is the native shape of the answer rather than a workaround — provided the workload scales horizontally.
| Region | AWS on-demand | AWS Spot | 1-yr Reserved | 3-yr Reserved | Rackspace Spot | You save |
|---|---|---|---|---|---|---|
| US East, N. Virginia | $7.657 | $1.486 | $5.046 | $3.365 | $0.492 | $7.165 savings |
| US Central, Chicago | no AWS region | $0.132 | ||||
| US Central, Dallas | no AWS region | $1.332 | ||||
| Hong Kong | $9.610 | $6.425 | $4.448 | $0.492 | $9.118 savings | |
| Sydney, Australia | $9.965 | $1.950 | $6.579 | $4.528 | $0.492 | $9.473 savings |
| London, UK | $9.066 | $1.207 | $5.985 | $4.120 | $1.572 | $7.494 savings |
| US West, San Jose | $9.515 | $1.390 | $6.284 | $4.327 | $0.024 | $9.492 savings |
| US Central, Dallas 2 | no AWS region | $1.464 | ||||
| US East, N. Virginia 2 | $7.657 | $1.486 | $5.046 | $3.365 | $1.944 | $5.713 savings |