Instances›AWS EC2›i8ge›i8ge.3xlarge
Compute
vCPUs12
Memory96 GiB
Mem / vCPU8.00 GiB
ProcessorGraviton4 Processor
Clock speed2.7 GHz
Architecturearm64
Storage
Type1 × 7,500 GB NVMe SSD
Bandwidth10,000 Mbps
Throughput1250 MB/s
IOPS40,000
Networking
PerformanceUp to 25 Gigabit
Max ENIs6
IPv6Yes
GPUNone
Family sizes
i8ge.large2 vCPU · 16 GiB
i8ge.xlarge4 vCPU · 32 GiB
i8ge.2xlarge8 vCPU · 64 GiB
i8ge.3xlarge12 vCPU · 96 GiB
i8ge.6xlarge24 vCPU · 192 GiB
i8ge.12xlarge48 vCPU · 384 GiB
i8ge.18xlarge72 vCPU · 576 GiB
i8ge.24xlarge96 vCPU · 768 GiB
i8ge.metal-24xl96 vCPU · 768 GiB
i8ge.48xlarge192 vCPU · 1536 GiB
i8ge.metal-48xl192 vCPU · 1536 GiB
i8ge.3xlarge

The i8ge.3xlarge is a storage-optimized AWS EC2 instance: 12 vCPU and 96 GiB of memory, 8.00 GiB per vCPU. In the name, "i" is I/O, meaning local storage, "8" is the generation, "ge" marks AWS Graviton (ARM) processors and extended capacity and "3xlarge" is the size within the i8ge family. It runs on AWS Graviton4 Processor at 2.7 GHz, and AWS introduced it in August 2025. i8ge is the current generation of the i family, so there is no newer AWS size to migrate to. The nearest Spot node is mh.vs1.2xlarge at 16 vCPU and 128 GiB — 4 more vCPU and 32 GiB more memory, $1.42/hr on AWS in US East (N. Virginia), from $0.021/hr on Rackspace Spot.

Match qualityx86 substitution

i8ge.3xlarge is Graviton (arm64). Rackspace Spot does not have arm64 servers, so mh.vs1.2xlarge is an x86_64 substitute with similar specs.

You will need x86 builds of your container images. Most software rebuilds without changes, but anything with compiled native code needs a real rebuild rather than a retag. Speed per core also differs between the two chips, so it is worth testing before deciding how many nodes to run.

No local diski8ge.3xlarge comes with 1 × 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$1.424$0.390$0.021
$1.403 savings
US Central, Chicagono AWS region$0.021
US Central, Dallasno AWS region$0.131
Hong Kong$1.881$0.021
$1.860 savings
Sydney, Australia$1.707$0.361$0.021
$1.686 savings
London, UK$1.657$0.292$0.131
$1.526 savings
US West, San Jose$0.002
US Central, Dallas 2no AWS region$0.242
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 96 GiB of RAM
  • ●Jobs that want 1 × 7,500 GB of local NVMe scratch space
  • ●Network-heavy services — AWS rates this size at Up to 25 Gigabit
Avoid if
  • ●The working set needs more than 96 GiB of RAM
  • ●You cannot rebuild x86-only binaries or vendor images for aarch64
  • ●More than 8.00 GiB of RAM per vCPU is needed — that ratio is fixed for the whole family
  • ●GPU or FPGA acceleration is needed
  • ●Workloads that never touch local disk. The instance-store premium is wasted.
$1,039/mo on AWS on-demand against $15/mo on Rackspace Spot for the 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