NVIDIA DGX A100 320GB

rack
8× A100
NVSwitch + NVLink
2020
FP64
0.1
PFLOPS
FP32
0.2
PFLOPS
Power
6.0
kW Total
Memory
320
GB Total

DGX A100 with 320GB total GPU memory

FP64
0.08
PFLOPS
12.9 TFLOPS/kW
FP32
0.16
PFLOPS
26.0 TFLOPS/kW
TF32
1.25
PFLOPS
208.0 TFLOPS/kW
FP16
2.50
PFLOPS
416.0 TFLOPS/kW

System Details

GPU Configuration

GPU Count: 8 GPUs
Architecture: Ampere
Interconnect: NVSwitch + NVLink

System Specifications

Form Factor: rack
Total Power: 6.0 kW
Total Memory: 320 GB
Memory Bandwidth: 12440 GB/s

Precision Performance Breakdown

PrecisionSystem PerformancePer GPUEfficiency
FP64 0.078 PFLOPS 9.7 TFLOPS 12.9 TFLOPS/kW
FP32 0.156 PFLOPS 19.5 TFLOPS 26.0 TFLOPS/kW
TF32 1.248 PFLOPS 156.0 TFLOPS 208.0 TFLOPS/kW
FP16 2.496 PFLOPS 312.0 TFLOPS 416.0 TFLOPS/kW
BF16 2.496 PFLOPS 312.0 TFLOPS 416.0 TFLOPS/kW
INT8 4.992 PFLOPS 624.0 TFLOPS 832.0 TFLOPS/kW

Powered by NVIDIA A100

This system utilizes 8 × NVIDIA A100 PCIe 40GB GPUs, each delivering exceptional performance for AI and HPC workloads.

Per GPU TDP

250W

Per GPU Memory

40 GB

Process Node

7nm

Architecture

Ampere

Documentation & Resources

workstation-datasheet-dgx-spark-gtc25-spring-nvidia-us-3716899-web.pdf

NVIDIA • 2025-10-16

View Document ↗

NVIDIA DGX Documentation

System guides, deployment resources

Browse ↗

Typical Use Cases

Large Language Model Training
Distributed Deep Learning
Multi-GPU Inference
HPC Simulations
Scientific Computing

The NVIDIA DGX A100 320GB with 8× A100 GPUs is designed for enterprise-scale workloads requiring 0.1 PFLOPS of compute power with NVSwitch + NVLink interconnect for efficient multi-GPU communication.