Instances›AWS EC2›i4i›i4i.32xlarge
Compute
vCPUs128
Memory1024 GiB
Mem / vCPU8.00 GiB
ProcessorXeon 8375C
Clock speed3.5 GHz
Architecturex86_64
Storage
Type8 × 3,750 GB NVMe SSD
Bandwidth40,000 Mbps
Throughput5000 MB/s
IOPS160,000
Networking
Performance75 Gigabit
Max ENIs15
IPv6Yes
GPUNone
Family sizes
i4i.large2 vCPU · 16 GiB
i4i.xlarge4 vCPU · 32 GiB
i4i.2xlarge8 vCPU · 64 GiB
i4i.4xlarge16 vCPU · 128 GiB
i4i.8xlarge32 vCPU · 256 GiB
i4i.12xlarge48 vCPU · 384 GiB
i4i.16xlarge64 vCPU · 512 GiB
i4i.24xlarge96 vCPU · 768 GiB
i4i.32xlarge128 vCPU · 1024 GiB
i4i.metal128 vCPU · 1024 GiB
i4i.32xlarge

i4i.32xlarge is AWS's storage-optimized EC2 size at 128 vCPU and 1024 GiB. In the name, "i" is I/O, meaning local storage, "4" is the generation, "i" marks Intel processors and "32xlarge" is the size within the i4i family. It runs on Intel Xeon 8375C (Ice Lake) at 3.5 GHz, and AWS introduced it in April 2022. AWS has since shipped i7i and i8g. The closest Spot shape is mh.vs1.2xlarge at 16 vCPU, so this instance maps to 8 of them — 128 vCPU and 1024 GiB across the fleet, $10.98/hr on AWS in US East (N. Virginia), from $0.168/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

i4i.32xlarge gives you 128 vCPU and 1,024 GiB in a single machine. The largest comparable Spot node, mh.vs1.2xlarge, is 16 vCPU and 128 GiB, so the equivalent here is 8 nodes rather than one.

Spot runs as managed Kubernetes, so a 8-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 128 cores and 1,024 GiB in one process, it does not.

No local diski4i.32xlarge comes with 8 × 3,750 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.
Pricing by region
1 hr
RegionAWS on-demandAWS Spot1-yr Reserved3-yr ReservedRackspace SpotYou save
US East, N. Virginia$10.982$2.971$7.127$5.080$0.168
$10.814 savings
US Central, Chicagono AWS region$0.168
US Central, Dallasno AWS region$1.048
Hong Kong$14.502$9.455$6.731$0.168
$14.334 savings
Sydney, Australia$13.165$2.144$8.594$6.096$0.168
$12.997 savings
London, UK$12.742$1.482$8.316$5.879$1.048
$11.694 savings
US West, San Jose$12.109$4.834$7.896$5.588$0.016
$12.093 savings
US Central, Dallas 2no AWS region$1.936
When to use it
Good for
  • ●Elasticsearch and OpenSearch data nodes
  • ●NoSQL stores such as Cassandra, ScyllaDB and MongoDB
  • ●Data warehousing and log processing
  • ●Workloads that fit inside 1024 GiB of RAM
  • ●Jobs that want 8 × 3,750 GB of local NVMe scratch space
  • ●Network-heavy services — AWS rates this size at 75 Gigabit
  • ●Anything that shards cleanly across 8 nodes, which is how it maps onto Rackspace Spot
Avoid if
  • ●The working set needs more than 1024 GiB of RAM
  • ●Workloads that never touch local disk. The instance-store premium is wasted.
  • ●More than 8.00 GiB of RAM per vCPU is needed — that ratio is fixed for the whole family
  • ●The work is single-threaded. Most of these 128 vCPU would sit idle.
  • ●GPU or FPGA acceleration is needed
i4i.32xlarge runs $8,017/mo on AWS on-demand. The 8-node equivalent on Rackspace Spot: $123/mo.
  • 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