p4de.24xlarge
Amazon EC2 · P4de family
The p4de.24xlarge is a gpu & accelerated Amazon EC2 instance type with 96 vCPUs and 1,152 GiB of memory, 8 × NVIDIA A100 80GB, running on Intel Xeon Platinum 8275L with 400 Gigabit networking, starting at $27.45 per hour on demand.
- vCPUs
- 96
- memory
- 1,152 GiB
- gpu
- 8× A100 80GB
- on-demand
- $27.45
- spot
- –
- 1-yr reserved
- $17.40
dedicated
12 GiB per vCPU
640 GiB total
$20,036/mo · us-east-1
not offered
−37% vs on-demand
p4de.24xlarge price by region
6 regions · hourly list prices in USD · On-demand · Spot · 1-yr Reserved · 3-yr Reserved
| US East (N. Virginia)us-east-1cheapestreferenceNorth America | $27.45 | $17.40 | $11.72 |
| US West (Oregon)us-west-2cheapestNorth America | $27.45 | $17.40 | $11.72 |
| Asia Pacific (Singapore)ap-southeast-1Asia | $32.94 | $20.88 | $14.06 |
| EU (Frankfurt)eu-central-1Europe | $34.29 | $21.74 | $14.64 |
| Israel (Tel Aviv)il-central-1Middle East | $35.57 | $22.55 | $15.19 |
| Asia Pacific (Tokyo)ap-northeast-1Asia | $37.62 | $24.47 | $19.74 |
Specifications
compute
- vCPUs
- 96
- Memory
- 1,152 GiB
- Memory per vCPU
- 12 GiB
- Processor
- Intel Xeon Platinum 8275L
- CPU vendor
- Intel
- Architecture
- x86-64
- Clock speed
- 3 GHz
- Burstable CPU
- No
storage & network
- Local storage
- 8 TB SSD (8 disks)
- Storage detail
- 8 x 1000 SSD
- Network
- 400 Gigabit
- Peak bandwidth
- 400 Gbps
accelerators
- Model
- NVIDIA A100 80GB
- Count
- 8
- Memory (total)
- 640 GiB
- Architecture
- Ampere
- Vendor
- NVIDIA
aws details
- API name
p4de.24xlarge- Family
- P4de
- Category
- GPU & accelerated
- Generation
- Current
- Regions
- 6 (Asia, Europe, Middle East, North America)
- Cheapest region
- US East (N. Virginia) · $27.45/hr
- Dedicated EBS throughput
- 19000 Mbps
- ECU
- 345
- Enhanced networking
- No
- Normalization size factor
- 192
- Processor features
- Intel AVX; Intel AVX2; Intel AVX512; Intel Turbo
- Spot interruption rate (us-east-1)
- 15-20%
- Spot savings vs on-demand (us-east-1)
- 37%
p4de.24xlarge equivalents on other clouds
closest match by vCPU, memory, CPU architecture, accelerator and class · Linux on-demand in each provider's reference region
| Provider | Instance | vCPU | Memory | Processor | On-demand | vs p4de.24xlarge |
|---|---|---|---|---|---|---|
| AWS | p4de.24xlarge | 96 | 1,152 GiB | 8 × NVIDIA A100 80GB | $27.45 | – |
| Azure | ND96amsr A100 v41900 GiB (1.65×) | 96 | 1,900 GiB | 8 × NVIDIA A100 80GB | $32.77 | +19% |
| NC96ads A100 v4880 GiB (1.3× less) · 4× GPU | 96 | 880 GiB | 4 × NVIDIA A100 80GB | $14.69 | -46% | |
| Google Cloud | a2-ultragpu-8g1360 GiB (1.18×) | 96 | 1,360 GiB | 8 × NVIDIA A100 80GB | $40.55 | +48% |
P4de family sizes
| Size | vCPU | Memory | Accelerator | Local storage | On-demand | Monthly | 1-yr |
|---|---|---|---|---|---|---|---|
| p4de.24xlarge | 96 | 1,152 GiB | 8 × NVIDIA A100 80GB | 8 TB SSD (8 disks) | $27.45 | $20,036 | $17.40 |
p4de.24xlarge FAQ
How much does p4de.24xlarge cost?
p4de.24xlarge costs $27.45 per hour on demand with Linux in US East (N. Virginia) (us-east-1), about $20,036 per month. Across 6 regions the Linux on-demand price ranges from $27.45 in US East (N. Virginia) to $37.62 in Asia Pacific (Tokyo). A 1-yr Reserved commitment brings it to $17.40 per hour.
How many vCPUs and how much memory does p4de.24xlarge have?
p4de.24xlarge has 96 vCPUs and 1,152 GiB of memory (12 GiB per vCPU), running on Intel Xeon Platinum 8275L. It is a gpu & accelerated instance type in the P4de family.
What GPU does p4de.24xlarge have?
p4de.24xlarge has 8 × NVIDIA A100 80GB with 640 GiB of accelerator memory in total.
What are the equivalents of p4de.24xlarge on other clouds?
The closest matches by vCPU, memory, CPU architecture, accelerator and instance class are on Azure, ND96amsr A100 v4 (96 vCPU, 1,900 GiB, $32.77/hour in East US); on Google Cloud, a2-ultragpu-8g (96 vCPU, 1,360 GiB, $40.55/hour in Iowa).