Instances›AWS EC2›m8azn›m8azn.3xlarge
Compute
vCPUs12
Memory48 GiB
Mem / vCPU4.00 GiB
ProcessorEPYC 9575F
Clock speedUp to 5 GHz
Architecturex86_64
Storage
TypeEBS only
Bandwidth15,000 Mbps
Throughput1875 MB/s
IOPS60,000
Networking
PerformanceUp to 50 Gigabit
Max ENIs8
IPv6Yes
GPUNone
Family sizes
m8azn.medium1 vCPU · 4 GiB
m8azn.large2 vCPU · 8 GiB
m8azn.xlarge4 vCPU · 16 GiB
m8azn.3xlarge12 vCPU · 48 GiB
m8azn.6xlarge24 vCPU · 96 GiB
m8azn.12xlarge48 vCPU · 192 GiB
m8azn.metal-12xl48 vCPU · 192 GiB
m8azn.24xlarge96 vCPU · 384 GiB
m8azn.metal-24xl96 vCPU · 384 GiB
m8azn.3xlarge

The m8azn.3xlarge is a general-purpose AWS EC2 instance: 12 vCPU and 48 GiB of memory, 4.00 GiB per vCPU. In the name, "m" is the general-purpose line, "8" is the generation, "azn" marks AMD processors, a higher clock and enhanced networking and "3xlarge" is the size within the m8azn family. It runs on AMD EPYC 9575F at Up to 5 GHz, and AWS introduced it in February 2026. AWS has since shipped m9g. The nearest Spot node is gp.vs2.2xlarge at 16 vCPU and 64 GiB — 4 more vCPU and 16 GiB more memory, $1.24/hr on AWS in US East (N. Virginia), from $0.041/hr on Rackspace Spot.

Pricing by region
1 hr
RegionAWS on-demandAWS Spot1-yr Reserved3-yr ReservedRackspace SpotYou save
US East, N. Virginia$1.239$0.417$0.041
$1.198 savings
US Central, Chicagono AWS region$0.041
US Central, Dallasno AWS region$0.051
Hong Kong$0.011
Sydney, Australia$0.161
London, UK$0.221
US West, San Jose$0.151
US Central, Dallas 2no AWS region$0.001
US East, N. Virginia 2$1.239$0.417$0.151
$1.088 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 48 GiB of RAM
  • ●Parallel work that can occupy 12 vCPU
  • ●Network-heavy services — AWS rates this size at Up to 50 Gigabit
Avoid if
  • ●The working set needs more than 48 GiB of RAM
  • ●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
  • ●Local NVMe or instance-store disk is a hard requirement
  • ●GPU or FPGA acceleration is needed
Stop overpaying AWS. This compute costs $904/mo on AWS vs $30/mo on Rackspace Spot.
  • 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