Instances›AWS EC2›d3en›d3en.8xlarge
Compute
vCPUs32
Memory128 GiB
Mem / vCPU4.00 GiB
ProcessorXeon Platinum 8259
Clock speed3.1 GHz
Architecturex86_64
Storage
Type16 × 13,980 GB NVMe SSD
Bandwidth5,000 Mbps
Throughput625 MB/s
IOPS30,000
Networking
Performance50 Gigabit
Max ENIs4
IPv6Yes
GPUNone
Family sizes
d3en.large2 vCPU · 8 GiB
d3en.xlarge4 vCPU · 16 GiB
d3en.2xlarge8 vCPU · 32 GiB
d3en.4xlarge16 vCPU · 64 GiB
d3en.6xlarge24 vCPU · 96 GiB
d3en.8xlarge32 vCPU · 128 GiB
d3en.12xlarge48 vCPU · 192 GiB
d3en.8xlarge

d3en.8xlarge is AWS's storage-optimized EC2 size at 32 vCPU and 128 GiB. In the name, "d" is dense local storage, "3" is the generation, "en" marks extended capacity and enhanced networking and "8xlarge" is the size within the d3en family. It runs on Intel Xeon Platinum 8259 (Cascade Lake) at 3.1 GHz, and AWS introduced it in December 2020. d3en is the current generation of the d family, so there is no newer AWS size to migrate to. The closest Spot shape is gp.vs1.2xlarge at 16 vCPU, so this instance maps to 2 of them — 32 vCPU and 128 GiB across the fleet, $4.21/hr on AWS in US East (N. Virginia), from $0.082/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

d3en.8xlarge gives you 32 vCPU and 128 GiB in a single machine. The largest comparable Spot node, gp.vs1.2xlarge, is 16 vCPU and 64 GiB, so the equivalent here is 2 nodes rather than one.

Spot runs as managed Kubernetes, so a 2-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 32 cores and 128 GiB in one process, it does not.

No local diskd3en.8xlarge comes with 16 × 13,980 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$4.206$0.853$2.650$1.817$0.082
$4.124 savings
US Central, Chicagono AWS region$0.082
US Central, Dallasno AWS region$0.102
Hong Kong$0.022
Sydney, Australia$5.272$0.901$3.321$2.277$0.322
$4.950 savings
London, UK$0.442
US West, San Jose$0.302
US Central, Dallas 2no AWS region$0.002
US East, N. Virginia 2$4.206$0.853$2.650$1.817$0.302
$3.904 savings
When to use it
Good for
  • ●Data warehousing and log processing
  • ●Distributed file systems and search indexes
  • ●Transactional databases with heavy random I/O
  • ●Workloads that fit inside 128 GiB of RAM
  • ●Jobs that want 16 × 13,980 GB of local NVMe scratch space
  • ●Network-heavy services — AWS rates this size at 50 Gigabit
  • ●Anything that shards cleanly across 2 nodes, which is how it maps onto Rackspace Spot
Avoid if
  • ●The working set needs more than 128 GiB of RAM
  • ●Workloads that never touch local disk. The instance-store premium is wasted.
  • ●More than 4.00 GiB of RAM per vCPU is needed — that ratio is fixed for the whole family
  • ●The work is single-threaded. Most of these 32 vCPU would sit idle.
  • ●GPU or FPGA acceleration is needed
$3,070/mo on AWS on-demand against $60/mo on Rackspace Spot for the 2-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