Instances›AWS EC2›hpc7a›hpc7a.96xlarge
Compute
vCPUs192
Memory768 GiB
Mem / vCPU4.00 GiB
ProcessorEPYC 9R14 Processor
Clock speed3.7 GHz
Architecturex86_64
Storage
TypeEBS only
Bandwidth2,085 Mbps
Throughput260.625 MB/s
IOPS11,000
Networking
Performance300 Gigabit
Max ENIs4
IPv6Yes
GPUNone
Family sizes
hpc7a.12xlarge24 vCPU · 768 GiB
hpc7a.24xlarge48 vCPU · 768 GiB
hpc7a.48xlarge96 vCPU · 768 GiB
hpc7a.96xlarge192 vCPU · 768 GiB
hpc7a.96xlarge

The hpc7a.96xlarge is a compute-optimized AWS EC2 instance: 192 vCPU and 768 GiB of memory, 4.00 GiB per vCPU. In the name, "hpc" is high performance computing, "7" is the generation, "a" marks AMD processors and "96xlarge" is the size within the hpc7a family. It runs on AMD EPYC 9R14 Processor at 3.7 GHz, and AWS introduced it in August 2023. AWS has since shipped hpc8a; hpc8a.96xlarge is the direct replacement at this size. Spot's largest VM is 16 vCPU and 128 GiB, so matching this instance would take 22 × ch.vs1.2xlarge — arithmetically true and operationally absurd. The Spot column below prices that whole fleet, so read it as a scale reference rather than a drop-in replacement: at this size the useful question is what a shard of the workload costs.

Match qualityNo single-node equivalent

hpc7a.96xlarge is 192 vCPU and 768 GiB. The largest comparable Spot node is ch.vs1.2xlarge at 16 vCPU and 32 GiB, so matching it would take 22 of them. That maths works out, but 22 nodes is a redesign rather than a migration — a single block of 768 GiB does not survive being split across a network.

If you chose hpc7a.96xlarge because the work genuinely needs 192 cores and 768 GiB together — an in-memory database, a large analytics engine, a single big solver — Rackspace Spot has no equivalent. What does fit is everything around it: the application servers, batch workers and CI that talk to that machine, which is usually the larger share of the bill.

Pricing by region
1 hr
RegionAWS on-demandAWS Spot1-yr Reserved3-yr ReservedRackspace SpotYou save
US East, N. Virginia$0.902
US Central, Chicagono AWS region$0.242
US Central, Dallasno AWS region$2.442
Hong Kong$0.902
Sydney, Australia$0.902
London, UK$2.882
US West, San Jose$0.044
US Central, Dallas 2no AWS region$2.684
US East, N. Virginia 2$3.564
When to use it
Good for
  • ●Video encoding and transcoding
  • ●High-traffic web and API serving
  • ●Scientific modelling and simulation
  • ●Workloads that fit inside 768 GiB of RAM
  • ●Parallel work that can occupy 192 vCPU
  • ●Network-heavy services — AWS rates this size at 300 Gigabit
Avoid if
  • ●The working set needs more than 768 GiB of RAM
  • ●You want a single-machine Spot equivalent. At this size there is not one.
  • ●The work is single-threaded. Most of these 192 vCPU would sit idle.
  • ●Local NVMe or instance-store disk is a hard requirement
  • ●GPU or FPGA acceleration is needed
  • ●Memory-hungry applications. A memory-optimized family gives more RAM per dollar.
Rackspace Spot sells spare capacity at market prices. You pay what the market decides, not a list price.
  • Spot sells spare capacity at market prices. It is the cheapest compute on the internet.
  • Prices are set by supply and demand. You pay what the market decides.
  • Compute is delivered as fully managed Kubernetes. All infrastructure is taken care of.
  • Egress costs $0.01/GB after the free allowance. AWS charges $0.09/GB.
Build your Cloudspace
Frequently asked questions