AWS M5d instance types

M5d is a general purpose Amazon EC2 family with 9 instance types, from 2 – 96 vCPUs and 8 GiB – 384 GiB of memory, running on Intel Xeon Platinum 8175. Linux on demand starts at $0.113 per hour.

general purposex86-64
sizes
9
vCPUs
2 – 96
memory
8 GiB – 384 GiB
processor
Intel Xeon Platinum 8175
on-demand from
$0.113

per hour · 31 regions

M5d sizes and prices

Linux, hourly, US East (N. Virginia) (other regions on each size's page)

SizevCPUMemoryLocal storageOn-demandMonthlySpot1-yr
m5d.large28 GiB75 GB NVME$0.113$82.49$0.0501$0.071
m5d.xlarge416 GiB150 GB NVME$0.226$165$0.121$0.142
m5d.2xlarge832 GiB300 GB NVME$0.452$330$0.2397$0.285
m5d.4xlarge1664 GiB600 GB NVME (2 disks)$0.904$660$0.4406$0.57
m5d.8xlarge32128 GiB1.2 TB NVME (2 disks)$1.808$1,320$0.9112$1.139
m5d.12xlarge48192 GiB1.8 TB NVME (2 disks)$2.712$1,980$1.145$1.709
m5d.16xlarge64256 GiB2.4 TB NVME (4 disks)$3.616$2,640$1.537$2.278
m5d.24xlarge96384 GiB3.6 TB NVME (4 disks)$5.424$3,960$2.456$3.417
m5d.metal96384 GiB3.6 TB NVME (4 disks)$5.424$3,960$2.093$3.417

M5d equivalents on other clouds

closest match to m5d.8xlarge (32 vCPU, 128 GiB)

AzureD32as v4Dasv4 family · 32 vCPU · 128 GiB$1.536/hr · East USDasv4 family →
Google Cloudc4d-standard-32-lssdC4D family · 32 vCPU · 124 GiB$1.676/hr · IowaC4D family →
AkamaiG8 Dedicated 128x32Dedicated CPU (G8) family · 32 vCPU · 128 GiB$1.68/hr · Newark, NJDedicated CPU (G8) family →
Vultrvx1-g-32c-128g-1920sVX1: General Purpose family · 32 vCPU · 128 GiB$1.223/hr · FrankfurtVX1: General Purpose family →

Other general purpose families on AWS

M5d FAQ

What sizes does the AWS M5d family have?

M5d has 9 instance types, from m5d.large (2 vCPU, 8 GiB) to m5d.metal (96 vCPU, 384 GiB). They run on Intel Xeon Platinum 8175.

What is the cheapest M5d instance type?

m5d.large at $0.113 per hour (Linux, on demand) in US East (N. Virginia), about $82.49 per month.

What is M5d best for?

M5d is a general purpose family. Balanced CPU and memory (about 4 GiB per vCPU) for web apps, services and most workloads.