The m6idn.24xlarge is a general-purpose AWS EC2 instance: 96 vCPU and 384 GiB of memory, 4.00 GiB per vCPU. In the name, "m" is the general-purpose line, "6" is the generation, "idn" marks Intel processors, local NVMe storage and enhanced networking and "24xlarge" is the size within the m6idn family. It runs on Intel Xeon 8375C (Ice Lake) at 3.5 GHz, and AWS introduced it in November 2022. AWS has since shipped m7i, m8idn and m9g; m7i.24xlarge is the direct replacement at this size, scoring 2,034,760 CoreMark/s against this instance's 1,646,655 — roughly 24% more throughput for the same vCPU count. No single Spot node covers 96 vCPU and 384 GiB, so the equivalent is a fleet: 6 × gp.vs2.2xlarge, 96 vCPU and 384 GiB in total, $7.64/hr on AWS in US East (N. Virginia), from $0.246/hr on Rackspace Spot. 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.
Match qualityScale-out match
m6idn.24xlarge gives you 96 vCPU and 384 GiB in a single machine. The largest comparable Spot node, gp.vs2.2xlarge, is 16 vCPU and 64 GiB, so the equivalent here is 6 nodes rather than one.
Spot runs as managed Kubernetes, so a 6-node pool is the normal way to run things rather than a workaround. If your workload already sits behind a load balancer, a queue or a scheduler, it spreads across them without changes. If it needs all 96 cores and 384 GiB in one process, it does not.
No local diskm6idn.24xlarge comes with 4 × 1,425 GB NVMe SSD attached to the machine. Spot nodes use network storage only, so that disk doesn't come with it. If you were using it as fast scratch space, you'd move to a persistent volume or object storage, which are slower to read.
