r7a.12xlarge

Amazon EC2 · R7a family

The r7a.12xlarge is a memory optimized Amazon EC2 instance type with 48 vCPUs and 384 GiB of memory, running on AMD EPYC 9R14 Processor with 18.75 Gigabit networking, starting at $3.652 per hour on demand.

memory optimizedx86-64amd
vCPUs
48

dedicated

memory
384 GiB

8 GiB per vCPU

processor
AMD

x86-64

on-demand
$3.652

$2,666/mo · us-east-1

spot
$1.717

−53% vs on-demand

1-yr reserved
$2.416

−34% vs on-demand

r7a.12xlarge price by region

9 regions · hourly list prices in USD · On-demand · Spot · 1-yr Reserved · 3-yr Reserved

US East (N. Virginia)us-east-1cheapestreferenceNorth America$3.652$1.717$2.416$1.656
US East (Ohio)us-east-2cheapestNorth America$3.652$1.195$2.416$1.656
US West (Oregon)us-west-2cheapestNorth America$3.652$1.052$2.416$1.656
Asia Pacific (Hyderabad)ap-south-2Asia$3.768–$2.48$1.709
EU (Stockholm)eu-north-1Europe$3.883$0.6486$2.569$1.761
EU (Ireland)eu-west-1Europe$4.086$1.583$2.703$1.854
EU (Spain)eu-south-2Europe$4.086$0.8177$2.703$1.854
Asia Pacific (Tokyo)ap-northeast-1Asia$4.405$1.173$2.914$1.998
EU (Frankfurt)eu-central-1Europe$4.405$1.416$2.914$1.998

Specifications

compute

vCPUs
48
Memory
384 GiB
Memory per vCPU
8 GiB
Processor
AMD EPYC 9R14 Processor
CPU vendor
AMD
Architecture
x86-64
Burstable CPU
No

storage & network

Local storage
Network storage only
Storage detail
EBS only
Network
18.75 Gigabit
Peak bandwidth
18.75 Gbps

aws details

API name
r7a.12xlarge
Family
R7a
Category
Memory optimized
Generation
Current
Regions
9 (Asia, Europe, North America)
Cheapest region
US East (N. Virginia) · $3.652/hr
Dedicated EBS throughput
15000 Mbps
Enhanced networking
Yes
Normalization size factor
96
Processor features
AMD Turbo; AVX; AVX2; AVX512
Spot interruption rate (us-east-1)
>20%
Spot savings vs on-demand (us-east-1)
70%

r7a.12xlarge equivalents on other clouds

closest match by vCPU, memory, CPU architecture, accelerator and class · Linux on-demand in each provider's reference region

compare side by side→
ProviderInstancevCPUMemoryProcessorOn-demandvs r7a.12xlarge
AWSr7a.12xlarge48384 GiBAMD EPYC 9R14 Processor$3.652–
AzureE48as v548384 GiBAMD EPYC 7763v (Milan)$2.712-26%
E48as v648384 GiBAMD EPYC 9004 (Genoa)$2.861-22%
Google Cloudn4d-highmem-4848384 GiBAMD Turin$2.668-27%
n2d-highmem-4848384 GiBAMD Milan, AMD Rome$2.736-25%
AkamaiDedicated 256GB56 vCPU (1.17×) · 256 GiB (1.5× less)56256 GiBAMD EPYC$3.456-5%
G7 Dedicated 256x5656 vCPU (1.17×) · 256 GiB (1.5× less)56256 GiBAMD EPYC$4.148+14%
Vultrvx1-m-48c-384gb48384 GiBAMD EPYC$1.92-47%
vx1-m-48c-384g-2880s48384 GiBAMD EPYC$2.315-37%

R7a family sizes

all about R7a→
SizevCPUMemoryLocal storageOn-demandMonthlySpot1-yr
r7a.medium18 GiB–$0.0761$55.54$0.0364$0.0503
r7a.large216 GiB–$0.1522$111$0.0698$0.1007
r7a.xlarge432 GiB–$0.3043$222$0.1408$0.2013
r7a.2xlarge864 GiB–$0.6086$444$0.2968$0.4026
r7a.4xlarge16128 GiB–$1.217$889$0.5987$0.8052
r7a.8xlarge32256 GiB–$2.434$1,777$0.9776$1.61
r7a.12xlarge48384 GiB–$3.652$2,666$1.717$2.416
r7a.16xlarge64512 GiB–$4.869$3,554$2.156$3.221
r7a.24xlarge96768 GiB–$7.303$5,331$2.993$4.831
r7a.32xlarge1281 TiB–$9.738$7,108$3.913$6.441
r7a.48xlarge1921,536 GiB–$14.61$10,663$6.077$9.662
r7a.metal-48xl1921,536 GiB–$14.61$10,663$4.778$9.662

Similar instance types on AWS

other AWS families with a similar shape, cheapest first

r7a.12xlarge FAQ

How much does r7a.12xlarge cost?

r7a.12xlarge costs $3.652 per hour on demand with Linux in US East (N. Virginia) (us-east-1), about $2,666 per month. Across 9 regions the Linux on-demand price ranges from $3.652 in US East (N. Virginia) to $4.405 in EU (Frankfurt). Spot capacity costs $1.717 per hour (53% less), but can be reclaimed. A 1-yr Reserved commitment brings it to $2.416 per hour. With Windows it costs $5.86 per hour.

How many vCPUs and how much memory does r7a.12xlarge have?

r7a.12xlarge has 48 vCPUs and 384 GiB of memory (8 GiB per vCPU), running on AMD EPYC 9R14 Processor. It is a memory optimized instance type in the R7a family.

What are the equivalents of r7a.12xlarge on other clouds?

The closest matches by vCPU, memory, CPU architecture and instance class are on Azure, E48as v5 (48 vCPU, 384 GiB, $2.712/hour in East US); on Google Cloud, n4d-highmem-48 (48 vCPU, 384 GiB, $2.668/hour in Iowa); on Akamai, Dedicated 256GB (56 vCPU, 256 GiB, $3.456/hour in Newark, NJ); on Vultr, vx1-m-48c-384gb (48 vCPU, 384 GiB, $1.92/hour in New Jersey).