r5d.24xlarge is AWS's memory-optimized EC2 size at 96 vCPU and 768 GiB. In the name, "r" stands for RAM, "5" is the generation, "d" marks local NVMe storage and "24xlarge" is the size within the r5d family. It runs on Intel Xeon Platinum 8175 at 3.1 GHz, and AWS introduced it in July 2018. AWS has since shipped r6i, r7i, r8i and r9g; r6i.24xlarge is the direct replacement at this size, scoring 1,705,908 CoreMark/s against this instance's 1,345,952 — roughly 27% more throughput for the same vCPU count. The closest Spot shape is mh.vs1.2xlarge at 16 vCPU, so this instance maps to 6 of them — 96 vCPU and 768 GiB across the fleet, $6.91/hr on AWS in US East (N. Virginia), from $0.126/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
r5d.24xlarge gives you 96 vCPU and 768 GiB in a single machine. The largest comparable Spot node, mh.vs1.2xlarge, is 16 vCPU and 128 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 768 GiB in one process, it does not.
No local diskr5d.24xlarge comes with 4 × 900 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.
