atlas
p GPU training
4 Generation 4
d Local NVMe
.
24xlarge 96 vCPU · 1152 GiB

$14,293/mo

1-yr reserved in Tokyo

96 vCPU

Xeon Cascade Lake, 3 GHz

1152 GiB

12 GiB per vCPU

8 TB SSD

Local storage

400 Gbps

Network bandwidth

19 Gbps

EBS bandwidth

x86_64

Architecture

Every way to pay

Prices in Tokyo. Bars are relative to on-demand.

PlanMonthlyYearlySaves

On-demand

Pay by the second, no commitment

$21,971$263,657

Reserved, 1 yr

No upfront, this size and region

$14,293$171,518−35%

Reserved, 3 yr

No upfront, this size and region

$11,526$138,316−48%

What will it cost me?

Enter how you plan to run it.

× p4d.24xlarge, on-demand

/mo

a year

You run these most of the time. A 1-year reservation bills every hour, and would still cost /mo.

Same usage on each planMonthlyYearly
On-demand
Reserved 1 yr
Reserved 3 yr

Reservations bill every hour of the month, whether the instance runs or not.

Other sizes in this family

Price roughly doubles with each step. Pick a size to open it.

Price in every region

Cheapest in US East (Ohio). Moving from Tokyo would save $49,569 a year per instance.

Not offered in N. California , Calgary , Central , US-East , Paris , Zurich , Milan , Spain , Hyderabad , Jakarta , Melbourne , Malaysia , New Zealand , Thailand , Osaka , Hong Kong , Taipei , UAE , Bahrain , Tel Aviv , Cape Town .

Same shape, other families

Every instance with 96 vCPU · 1152 GiB in Tokyo, ranked by price.

InstanceProcessorGenerationLocal diskPrice/movs p4d.24xlarge
p4d.24xlargeXeon Cascade LakeGen 48 TB SSD$14,293—
p4de.24xlargeXeon Cascade LakeGen 48 TB SSD$17,866+25%

Specifications

API name
p4d.24xlarge
Processor
Intel Xeon Platinum 8275L
Local storage
8 × 1 TB SSD
Family
GPU instance
Clock speed
3 GHz
Network
400 Gbps
Generation
Gen 4
Architecture
x86_64
EBS bandwidth
19 Gbps
vCPUs
96
Memory
1152 GiB
GPUs
8 · 320 GB HBM2