Instances›AWS EC2›m8in›m8in.metal-96xl
Compute
vCPUs384
Memory1536 GiB
Mem / vCPU4.00 GiB
ProcessorXeon Scalable
Clock speed3.9 GHz
Architecturex86_64
Storage
TypeEBS only
Bandwidth120,000 Mbps
Throughput15000 MB/s
IOPS480,000
Networking
Performance600 Gigabit
Max ENIs24
IPv6Yes
GPUNone
Family sizes
m8in.large2 vCPU · 8 GiB
m8in.xlarge4 vCPU · 16 GiB
m8in.2xlarge8 vCPU · 32 GiB
m8in.4xlarge16 vCPU · 64 GiB
m8in.8xlarge32 vCPU · 128 GiB
m8in.12xlarge48 vCPU · 192 GiB
m8in.16xlarge64 vCPU · 256 GiB
m8in.24xlarge96 vCPU · 384 GiB
m8in.32xlarge128 vCPU · 512 GiB
m8in.48xlarge192 vCPU · 768 GiB
m8in.metal-48xl192 vCPU · 768 GiB
m8in.96xlarge384 vCPU · 1536 GiB
m8in.metal-96xl384 vCPU · 1536 GiB
m8in.metal-96xl

m8in.metal-96xl is AWS's general-purpose EC2 size at 384 vCPU and 1536 GiB. In the name, "m" is the general-purpose line, "8" is the generation, "in" marks Intel processors and enhanced networking and "metal-96xl" is the size within the m8in family. It runs on Intel Xeon Scalable (Granite Rapids) at 3.9 GHz, and AWS introduced it in June 2026. AWS has since shipped m9g. Spot's largest VM is 16 vCPU and 128 GiB, so matching this instance would take 24 × gp.vs2.2xlarge — arithmetically true and operationally absurd. The Spot column below prices that whole fleet, so read it as a scale reference rather than a drop-in replacement: at this size the useful question is what a shard of the workload costs, which is $0.041/hr per node.

Match qualityNo single-node equivalent

m8in.metal-96xl is 384 vCPU and 1,536 GiB. The largest comparable Spot node is gp.vs2.2xlarge at 16 vCPU and 64 GiB, so matching it would take 24 of them. That maths works out, but 24 nodes is a redesign rather than a migration — a single block of 1,536 GiB does not survive being split across a network.

If you chose m8in.metal-96xl because the work genuinely needs 384 cores and 1,536 GiB together — an in-memory database, a large analytics engine, a single big solver — Rackspace Spot has no equivalent. What does fit is everything around it: the application servers, batch workers and CI that talk to that machine, which is usually the larger share of the bill.

Pricing by region
1 hr
RegionAWS on-demandAWS Spot1-yr Reserved3-yr ReservedRackspace SpotYou save
US East, N. Virginia$32.079$3.208$0.984
$31.095 savings
US Central, Chicagono AWS region$0.984
US Central, Dallasno AWS region$1.224
Hong Kong$0.264
Sydney, Australia$3.864
London, UK$5.304
US West, San Jose$3.624
US Central, Dallas 2no AWS region$0.024
US East, N. Virginia 2$32.079$3.208$3.624
$28.455 savings
When to use it
Good for
  • ●Queue workers and scheduled jobs
  • ●Web and application servers with mixed CPU and memory demand
  • ●Microservices and container fleets
  • ●Workloads that fit inside 1536 GiB of RAM
  • ●Parallel work that can occupy 384 vCPU
  • ●Network-heavy services — AWS rates this size at 600 Gigabit
Avoid if
  • ●The working set needs more than 1536 GiB of RAM
  • ●You want a single-machine Spot equivalent. At this size there is not one.
  • ●Sustained full-core CPU load. A compute-optimized family costs less per unit of work.
  • ●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 384 vCPU would sit idle.
  • ●Local NVMe or instance-store disk is a hard requirement
m8in.metal-96xl is $23,418/mo on AWS on-demand. A gp.vs2.2xlarge node on Rackspace Spot is $30/mo — at this size the per-node price is the honest comparison.
  • 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