Instances›AWS EC2›r3›r3.8xlarge
Compute
vCPUs32
Memory244 GiB
Mem / vCPU7.63 GiB
ProcessorXeon E5-2670 v2
Clock speed2.5 GHz
Architecturex86_64
Storage
Type2 × 320 GB SSD
BandwidthN/A
ThroughputN/A
IOPS0
Networking
Performance10 Gigabit
Max ENIs8
IPv6Yes
GPUNone
Family sizes
r3.large2 vCPU · 15.25 GiB
r3.xlarge4 vCPU · 30.5 GiB
r3.2xlarge8 vCPU · 61 GiB
r3.4xlarge16 vCPU · 122 GiB
r3.8xlarge32 vCPU · 244 GiB
r3.8xlarge

r3.8xlarge is AWS's memory-optimized EC2 size at 32 vCPU and 244 GiB. In the name, "r" stands for RAM, "3" is the generation and "8xlarge" is the size within the r3 family. It runs on Intel Xeon E5-2670 v2 (Ivy Bridge) at 2.5 GHz, and AWS introduced it in April 2014. AWS has since shipped r4, r5, r6i and r7i; r4.8xlarge is the direct replacement at this size. The closest Spot shape is mh.vs1.2xlarge at 16 vCPU, so this instance maps to 2 of them — 32 vCPU and 256 GiB across the fleet, $2.66/hr on AWS in US East (N. Virginia), from $0.042/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

r3.8xlarge gives you 32 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 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 244 GiB in one process, it does not.

No local diskr3.8xlarge comes with 2 × 320 GB 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$2.660$0.693$1.672$0.042
$2.618 savings
US Central, Chicagono AWS region$0.042
US Central, Dallasno AWS region$0.262
Hong Kong$0.042
Sydney, Australia$3.192$0.863$2.386$0.042
$3.150 savings
London, UK$0.262
US West, San Jose$2.964$0.593$2.174$0.004
$2.960 savings
US Central, Dallas 2no AWS region$0.484
When to use it
Good for
  • ●In-memory caches such as Redis and Memcached
  • ●Large JVM heaps and application servers that hold state
  • ●OLAP and business intelligence query engines
  • ●Workloads that fit inside 244 GiB of RAM
  • ●Jobs that want 2 × 320 GB of local SSD scratch space
  • ●Network-heavy services — AWS rates this size at 10 Gigabit
  • ●Anything that shards cleanly across 2 nodes, which is how it maps onto Rackspace Spot
Avoid if
  • ●Starting a new deployment. r4.8xlarge is the newer generation at this size.
  • ●The working set needs more than 244 GiB of RAM
  • ●GPU or FPGA acceleration is needed
  • ●CPU-bound work. You are paying for memory this workload will not use.
  • ●More than 7.63 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.
r3.8xlarge runs $1,942/mo on AWS on-demand. The 2-node equivalent on Rackspace Spot: $31/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