Instances›AWS EC2›m8gn›m8gn.metal-24xl
Compute
vCPUs96
Memory384 GiB
Mem / vCPU4.00 GiB
ProcessorGraviton4 Processor
Clock speed2.8 GHz
Architecturearm64
Storage
TypeEBS only
Bandwidth60,000 Mbps
Throughput7500 MB/s
IOPS240,000
Networking
Performance300 Gigabit
Max ENIs24
IPv6Yes
GPUNone
Family sizes
m8gn.medium1 vCPU · 4 GiB
m8gn.large2 vCPU · 8 GiB
m8gn.xlarge4 vCPU · 16 GiB
m8gn.2xlarge8 vCPU · 32 GiB
m8gn.4xlarge16 vCPU · 64 GiB
m8gn.8xlarge32 vCPU · 128 GiB
m8gn.12xlarge48 vCPU · 192 GiB
m8gn.16xlarge64 vCPU · 256 GiB
m8gn.24xlarge96 vCPU · 384 GiB
m8gn.metal-24xl96 vCPU · 384 GiB
m8gn.48xlarge192 vCPU · 768 GiB
m8gn.metal-48xl192 vCPU · 768 GiB
m8gn.metal-24xl

m8gn.metal-24xl is AWS's general-purpose EC2 size at 96 vCPU and 384 GiB. In the name, "m" is the general-purpose line, "8" is the generation, "gn" marks AWS Graviton (ARM) processors and enhanced networking and "metal-24xl" is the size within the m8gn family. It runs on AWS Graviton4 Processor at 2.8 GHz, and AWS introduced it in February 2026. AWS has since shipped m9g. The closest Spot shape is gp.vs2.2xlarge at 16 vCPU, so this instance maps to 6 of them — 96 vCPU and 384 GiB across the fleet, $6.98/hr on AWS in US East (N. Virginia), from $0.246/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

m8gn.metal-24xl gives you 96 vCPU and 384 GiB in a single machine. The largest comparable Spot node, gp.vs2.2xlarge, is 16 vCPU and 64 GiB, so the equivalent here is 6 nodes rather than one.

Spot runs as managed Kubernetes, so a 6-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 96 cores and 384 GiB in one process, it does not.

x86_64 onlym8gn.metal-24xl is Graviton (arm64) and Rackspace Spot has no arm64 nodes, so you'd be moving to x86. You'll need x86 builds of your images, and per-core speed differs between the chips — worth testing before you decide how many nodes to run.
Pricing by region
1 hr
RegionAWS on-demandAWS Spot1-yr Reserved3-yr ReservedRackspace SpotYou save
US East, N. Virginia$6.984$1.049$0.246
$6.738 savings
US Central, Chicagono AWS region$0.246
US Central, Dallasno AWS region$0.306
Hong Kong$0.066
Sydney, Australia$0.966
London, UK$1.326
US West, San Jose$0.906
US Central, Dallas 2no AWS region$0.006
US East, N. Virginia 2$6.984$1.049$0.906
$6.078 savings
When to use it
Good for
  • ●Small to mid-size relational databases
  • ●Backend services with no single dominant resource
  • ●Queue workers and scheduled jobs
  • ●Workloads that fit inside 384 GiB of RAM
  • ●Parallel work that can occupy 96 vCPU
  • ●Network-heavy services — AWS rates this size at 300 Gigabit
  • ●Anything that shards cleanly across 6 nodes, which is how it maps onto Rackspace Spot
Avoid if
  • ●The working set needs more than 384 GiB of RAM
  • ●You cannot rebuild x86-only binaries or vendor images for aarch64
  • ●GPU or FPGA acceleration is needed
  • ●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 96 vCPU would sit idle.
$5,098/mo on AWS on-demand against $180/mo on Rackspace Spot for the 6-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