p4de.24xlarge by Amazon Web Services
(All-cores)
(Single-core)
Specifications
Server Metadata
Vendor ID | aws |
Name | p4de.24xlarge |
Description | GPU accelerated [Instance store volumes] [Extra storage or memory] Gen4 24xlarge |
Family | p4de |
Hw Virt | - |
Status | active |
Observed At | 2026-07-25T07:02:16.655370 |
Availability
| REGION / ID | SPOT | ONDEMAND |
|---|---|---|
Northern Virgina (US) / us-east-1 | 21.01095 USD/h | 27.4471 USD/h |
Oregon (US) / us-west-2 | 21.724733 USD/h | 27.4471 USD/h |
Singapore (SG) / ap-southeast-1 | 32.9365 USD/h | 32.9365 USD/h |
Frankfurt (DE) / eu-central-1 | 12.4348 USD/h | 34.2912 USD/h |
Tel Aviv (IL) / il-central-1 | 8.0641 USD/h | 35.5735 USD/h |
Tokyo (JP) / ap-northeast-1 | 21.62295 USD/h | 37.6223 USD/h |
Processor
vCPUs | 96 |
Hypervisor | nitro |
CPU Allocation | Dedicated |
CPU Cores | 48 |
CPU Speed | 3 GHz |
CPU Architecture | x86_64 |
CPU Manufacturer | Intel |
System Resources and Accelerators
| MEMORY | |
|---|---|
Memory Amount | 1 TB |
| GPU | |
|---|---|
GPU Count | 8 |
GPU Memory Min | 80 GiB |
GPU Memory Total | 640 GiB |
GPU Manufacturer | NVIDIA |
GPU Family | Ampere |
GPU Model | A100 |
GPUs |
|
| STORAGE | |
|---|---|
Storage Size | 8000 GB |
Storage Type | nvme ssd |
Storages |
|
| NETWORK | |
|---|---|
Network Speed Baseline | 100 Gbps |
Network Speed Max | 100 Gbps |
Network Storage Speed Baseline | 19 Gbps |
Network Storage Speed Max | 19 Gbps |
Inbound Traffic | 0 GB/month |
Outbound Traffic | 0 GB/month |
IPv4 | 0 |
Server Description
A high-density GPU instance featuring eight NVIDIA A100 accelerators, massive system memory, and high-speed local NVMe storage.
Amazon Web Services p4de.24xlarge is a GPU-accelerated instance designed for intensive parallel computing workloads. Built on the AWS Nitro hypervisor, it features 96 vCPUs powered by dedicated 3.0 GHz Intel processors on the x86_64 architecture, alongside 1152.0 GB of system memory. The hardware configuration includes eight NVIDIA Ampere A100 GPUs with a combined 640 GB of VRAM, providing substantial acceleration for mathematical computations. Local storage is supplied via 8000 GB of NVMe SSD instance store volumes, and network performance is anchored by a 100 Gbps baseline bandwidth. This high-density resource profile is engineered for deep learning training, large-scale artificial intelligence inference, and complex scientific simulations. It offers a balanced ratio of compute, memory, and storage resources to prevent bottlenecks during intensive data processing tasks.
Economics
Alternatives
Servers of the Same Family
| INSTANCE | vCPUs | MEMORY | GPUs |
|---|
Similar Servers
| INSTANCE | VENDOR | vCPUs | MEMORY | GPUs |
|---|