Instances›AWS EC2›hpc7g›hpc7g.16xlarge
Compute
vCPUs64
Memory128 GiB
Mem / vCPU2.00 GiB
ProcessorGraviton3 Processor
Clock speed2.6 GHz
Architecturearm64
Storage
TypeEBS only
Bandwidth2,085 Mbps
Throughput260.625 MB/s
IOPS11,000
Networking
Performance200 Gigabit
Max ENIs4
IPv6Yes
GPUNone
Family sizes
hpc7g.4xlarge16 vCPU · 128 GiB
hpc7g.8xlarge32 vCPU · 128 GiB
hpc7g.16xlarge64 vCPU · 128 GiB
hpc7g.16xlarge

The hpc7g.16xlarge is a compute-optimized AWS EC2 instance: 64 vCPU and 128 GiB of memory, 2.00 GiB per vCPU. In the name, "hpc" is high performance computing, "7" is the generation, "g" marks AWS Graviton (ARM) processors and "16xlarge" is the size within the hpc7g family. It runs on AWS Graviton3 Processor at 2.6 GHz, and AWS introduced it in June 2023. AWS has since shipped hpc8a. No single Spot node covers 64 vCPU and 128 GiB, so the equivalent is a fleet: 4 × ch.vs1.2xlarge, 64 vCPU and 128 GiB in total, about 90% 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.

Match qualityScale-out match

hpc7g.16xlarge gives you 64 vCPU and 128 GiB in a single machine. The largest comparable Spot node, ch.vs1.2xlarge, is 16 vCPU and 32 GiB, so the equivalent here is 4 nodes rather than one.

Spot runs as managed Kubernetes, so a 4-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 64 cores and 128 GiB in one process, it does not.

x86_64 onlyhpc7g.16xlarge is Graviton (arm64) and Rackspace Spot has no arm64 nodes, so you'd be moving to x86. You'll need x86 builds of your images, and per-core speed differs between the chips — worth testing before you decide how many nodes to run.
Pricing by region
1 hr
RegionAWS on-demandAWS Spot1-yr Reserved3-yr ReservedRackspace SpotYou save
US East, N. Virginia$1.683$1.325$1.163$0.164
$1.519 savings
US Central, Chicagono AWS region$0.044
US Central, Dallasno AWS region$0.444
Hong Kong$0.164
Sydney, Australia$0.164
London, UK$0.524
US West, San Jose$0.008
US Central, Dallas 2no AWS region$0.488
US East, N. Virginia 2$1.683$1.325$1.163$0.648
$1.035 savings
When to use it
Good for
  • ●High-traffic web and API serving
  • ●Scientific modelling and simulation
  • ●Ad serving and real-time bidding
  • ●Workloads that fit inside 128 GiB of RAM
  • ●Parallel work that can occupy 64 vCPU
  • ●Network-heavy services — AWS rates this size at 200 Gigabit
  • ●Anything that shards cleanly across 4 nodes, which is how it maps onto Rackspace Spot
Avoid if
  • ●The working set needs more than 128 GiB of RAM
  • ●You cannot rebuild x86-only binaries or vendor images for aarch64
  • ●The work is single-threaded. Most of these 64 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.
$1,229/mo on AWS on-demand against $120/mo on Rackspace Spot for the 4-node equivalent — market-priced, not list-priced.
  • 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