Instances›AWS EC2›i3en›i3en.12xlarge
Compute
vCPUs48
Memory384 GiB
Mem / vCPU8.00 GiB
ProcessorXeon Platinum 8175
Clock speed3.1 GHz
Architecturex86_64
Storage
Type4 × 7,500 GB NVMe SSD
Bandwidth9,500 Mbps
Throughput1187.5 MB/s
IOPS40,000
Networking
Performance50 Gigabit
Max ENIs8
IPv6Yes
GPUNone
Family sizes
i3en.large2 vCPU · 16 GiB
i3en.xlarge4 vCPU · 32 GiB
i3en.2xlarge8 vCPU · 64 GiB
i3en.3xlarge12 vCPU · 96 GiB
i3en.6xlarge24 vCPU · 192 GiB
i3en.12xlarge48 vCPU · 384 GiB
i3en.24xlarge96 vCPU · 768 GiB
i3en.metal96 vCPU · 768 GiB
i3en.12xlarge

i3en.12xlarge is AWS's storage-optimized EC2 size at 48 vCPU and 384 GiB. In the name, "i" is I/O, meaning local storage, "3" is the generation, "en" marks extended capacity and enhanced networking and "12xlarge" is the size within the i3en family. It runs on Intel Xeon Platinum 8175 at 3.1 GHz, and AWS introduced it in May 2019. AWS has since shipped i4i, i7i and i8g; i4i.12xlarge is the direct replacement at this size, scoring 817,926 CoreMark/s against this instance's 700,662 — roughly 17% more throughput for the same vCPU count. The closest Spot shape is mh.vs1.2xlarge at 16 vCPU, so this instance maps to 3 of them — 48 vCPU and 384 GiB across the fleet, $5.42/hr on AWS in US East (N. Virginia), from $0.063/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

i3en.12xlarge gives you 48 vCPU and 384 GiB in a single machine. The largest comparable Spot node, mh.vs1.2xlarge, is 16 vCPU and 128 GiB, so the equivalent here is 3 nodes rather than one.

Spot runs as managed Kubernetes, so a 3-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 48 cores and 384 GiB in one process, it does not.

No local diski3en.12xlarge comes with 4 × 7,500 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$5.424$1.714$3.694$2.360$0.063
$5.361 savings
US Central, Chicagono AWS region$0.063
US Central, Dallasno AWS region$0.393
Hong Kong$7.160$4.896$3.117$0.063
$7.097 savings
Sydney, Australia$6.504$1.361$4.448$2.831$0.063
$6.441 savings
London, UK$6.312$1.051$4.315$2.743$0.393
$5.919 savings
US West, San Jose$6.000$0.736$4.098$2.608$0.006
$5.994 savings
US Central, Dallas 2no AWS region$0.726
When to use it
Good for
  • ●Transactional databases with heavy random I/O
  • ●Streaming ingest that spools to local disk
  • ●Elasticsearch and OpenSearch data nodes
  • ●Workloads that fit inside 384 GiB of RAM
  • ●Jobs that want 4 × 7,500 GB of local NVMe scratch space
  • ●Network-heavy services — AWS rates this size at 50 Gigabit
  • ●Anything that shards cleanly across 3 nodes, which is how it maps onto Rackspace Spot
Avoid if
  • ●Starting a new deployment. i4i.12xlarge is the newer generation at this size.
  • ●The working set needs more than 384 GiB of RAM
  • ●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 48 vCPU would sit idle.
  • ●GPU or FPGA acceleration is needed
  • ●Workloads that never touch local disk. The instance-store premium is wasted.
i3en.12xlarge runs $3,960/mo on AWS on-demand. The 3-node equivalent on Rackspace Spot: $46/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