Instances›AWS EC2›d2›d2.8xlarge
Compute
vCPUs36
Memory244 GiB
Mem / vCPU6.78 GiB
ProcessorXeon E5-2676 v3
Clock speed2.4 GHz
Architecturex86_64
Storage
Type24 × 2,048 GB HDD
Bandwidth4,000 Mbps
Throughput500 MB/s
IOPS32,000
Networking
Performance10 Gigabit
Max ENIs8
IPv6Yes
GPUNone
Family sizes
d2.xlarge4 vCPU · 30.5 GiB
d2.2xlarge8 vCPU · 61 GiB
d2.4xlarge16 vCPU · 122 GiB
d2.8xlarge36 vCPU · 244 GiB
d2.8xlarge

The d2.8xlarge is a storage-optimized AWS EC2 instance: 36 vCPU and 244 GiB of memory, 6.78 GiB per vCPU. In the name, "d" is dense local storage, "2" is the generation and "8xlarge" is the size within the d2 family. It runs on Intel Xeon E5-2676 v3 (Haswell) at 2.4 GHz, and AWS introduced it in March 2015. AWS has since shipped d3; d3.8xlarge is the direct replacement at this size. No single Spot node covers 36 vCPU and 244 GiB, so the equivalent is a fleet: 3 × mh.vs1.2xlarge, 48 vCPU and 384 GiB in total, $5.52/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

d2.8xlarge gives you 36 vCPU and 244 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 36 cores and 244 GiB in one process, it does not.

No local diskd2.8xlarge comes with 24 × 2,048 GB HDD 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$5.520$1.562$3.216$0.063
$5.457 savings
US Central, Chicagono AWS region$0.063
US Central, Dallasno AWS region$0.393
Hong Kong$7.656$4.019$2.968$0.063
$7.593 savings
Sydney, Australia$6.960$1.377$4.104$0.063
$6.897 savings
London, UK$6.174$0.658$3.753$0.393
$5.781 savings
US West, San Jose$6.250$0.774$3.931$0.005
$6.245 savings
US Central, Dallas 2no AWS region$0.605
When to use it
Good for
  • ●NoSQL stores such as Cassandra, ScyllaDB and MongoDB
  • ●Data warehousing and log processing
  • ●Distributed file systems and search indexes
  • ●Workloads that fit inside 244 GiB of RAM
  • ●Jobs that want 24 × 2,048 GB of local SSD scratch space
  • ●Network-heavy services — AWS rates this size at 10 Gigabit
  • ●Anything that shards cleanly across 3 nodes, which is how it maps onto Rackspace Spot
Avoid if
  • ●Starting a new deployment. d3.8xlarge is the newer generation at this size.
  • ●The working set needs more than 244 GiB of RAM
  • ●More than 6.78 GiB of RAM per vCPU is needed — that ratio is fixed for the whole family
  • ●The work is single-threaded. Most of these 36 vCPU would sit idle.
  • ●GPU or FPGA acceleration is needed
  • ●Workloads that never touch local disk. The instance-store premium is wasted.
d2.8xlarge runs $4,030/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