AWS's p3dn.24xlarge is a GPU instance with 96 vCPU and 768 GiB. Rackspace Spot has no equivalent to offer: 8 × NVIDIA Tesla V100 wired together behind a high-speed interconnect is one tightly coupled machine, not 8 independent nodes, and that interconnect is usually what the instance was bought for.
Rackspace Spot's largest VM tops out at 16 vCPU and 128 GiB, and even that is not the missing piece here. The table below prices what AWS charges; the honest answer on the Rackspace Spot side is that this workload stays where it is.
Regions are Rackspace Spot locations.
| Region | AWS on-demand | AWS Spot | 1-yr Reserved | 3-yr Reserved | Rackspace on-demand | Rackspace Spot | On-demand price savings | Spot price savings |
|---|---|---|---|---|---|---|---|---|
| US East, N. Virginia | $31.212 | $3.121 | $19.215 | $10.416 | not offered | not offered | n/a | n/a |
| US Central, Chicago | no AWS region | n/a | n/a | n/a | not offered | not offered | n/a | n/a |
| US Central, Dallas | no AWS region | n/a | n/a | n/a | not offered | not offered | n/a | n/a |
| Hong Kong | not sold here | n/a | n/a | n/a | not offered | not offered | n/a | n/a |
| Sydney, Australia | not sold here | n/a | n/a | n/a | not offered | not offered | n/a | n/a |
| London, UK | not sold here | n/a | n/a | n/a | not offered | not offered | n/a | n/a |
| US West, San Jose | not sold here | n/a | n/a | n/a | not offered | not offered | n/a | n/a |
| US Central, Dallas 2 | no AWS region | n/a | n/a | n/a | not offered | not offered | n/a | n/a |
| US East, N. Virginia 2 | $31.212 | $3.121 | $19.215 | $10.416 | not offered | not offered | n/a | n/a |
Prices updated 18 Nov 2026.
The AWS p3dn.24xlarge costs $31.21/hr on-demand in US East (N. Virginia), about $22,785/mo running 24/7, or $3.12/hr at the AWS Spot price. Rackspace Spot also sells in Chicago and Dallas, where AWS operates no region at all, so those rows have no AWS price to compare against.
At 730 hours a month, p3dn.24xlarge in US East (N. Virginia) is $22,785/mo on AWS on-demand, or $2,278/mo at the AWS Spot price.
No, and it is worth saying so plainly: 8 × NVIDIA Tesla V100 wired together behind a high-speed interconnect is one tightly coupled machine, not 8 independent nodes, and that interconnect is usually what the instance was bought for. Rackspace Spot sells market-priced CPU and GPU capacity, which is a good fit for the batch, web and inference work around this instance, but not for p3dn.24xlarge itself.
For workloads already running on it, yes: it works and it is priced accordingly. For anything new, p4d.24xlarge is the current answer and measures roughly 20% more CoreMark throughput at the same vCPU count. 2 newer generations of the p family have shipped since (p4d and p5), which is the signal that matters for infrastructure you expect to keep running.
p4d is the newer generation. It runs on Intel Xeon Platinum 8275L against p3dn's Intel Xeon Platinum 8175 (Skylake). On the CoreMark figures in this dataset, p4d.24xlarge scores 1,593,361 per second against p3dn.24xlarge's 1,323,529. On-demand in US East (N. Virginia) the newer size is 30% cheaper. Neither generation changes the Rackspace Spot side: both map to the same Rackspace Spot servers.
Not on Rackspace Spot, for the reason above. Within AWS the usual levers are a smaller size in the same family, a reserved or savings-plan commitment, or AWS Spot capacity if the work can tolerate interruption.