NVIDIA A30 Professional Graphics Card

Product status: Official | Last Update: 2022-11-08 | Report Error
Overview
Manufacturer
NVIDIA
Original Series
Data Center Ampere
Launch Date
April 12th, 2021
P/N Code
699-21001-0205
Device ID
20B7
Model
NVIDIA P1001 SKU 205
Graphics Processing Unit
GPU Model
GA100
Architecture
Ampere
Fabrication Process
7 nm
Die Size
826 mm2
Transistors Count
54.2B
Transistors Density
65.6M TRAN/mm2
CUDAs
3584
Tensor Cores
224
SM
56
Clocks
Base Clock
930 MHz
Boost Clock
1440 MHz
Memory Clock
1215 MHz
Effective Memory Clock
2430 Mbps
Memory Configuration
Memory Size
24576 MB
Memory Type
HBM2
Memory Bus Width
3072-bit
Memory Bandwidth
933.1 GB/s

Physical
Interface
PCI-Express 4.0 x16
Width
11.15 cm
Length
26.86 cm
Height
1-slot
Power Connectors
1× 8-pin
TDP/TBP
165 W
Recommended PSU
500 W
Multi-GPU Support
Unsupported
API Support
DirectX
12.2
Vulkan
1.2
OpenGL
4.6
OpenCL
2.1

Performance
Peak FP32
10.3 TFLOPS
Peak FP64
5.2 TFLOPS
FP32 Perf. per Watt
62.6 GFLOPS/W
FP32 Perf. per mm2
12.5 GFLOPS/mm2




 ModelCoresBoost ClockMemory ClockMemory Config.
Thumbnail
NVIDIA HGX A100 (16-GPU)
 
110592
 
1410 MHz
 
2.4 Gbps
 
10240 GB HB2 5120b
Thumbnail
NVIDIA DGX A100
 
55296
 
1410 MHz
 
2.4 Gbps
 
2560 GB HB2 5120b
Thumbnail
NVIDIA HGX A100 (8-GPU)
 
55296
 
1410 MHz
 
2.4 Gbps
 
2560 GB HB2 5120b
Thumbnail
NVIDIA HGX A100 (4-GPU)
 
27648
 
1410 MHz
 
2.4 Gbps
 
640 GB HB2 5120b
Thumbnail
NVIDIA A16
 
5120
 
1755 MHz
 
12.5 Gbps
 
256 GB G6 128b
Thumbnail
NVIDIA A40
 
10752
 
1740 MHz
 
14.5 Gbps
 
48 GB G6 384b
Thumbnail
NVIDIA A10
 
9216
 
1695 MHz
 
12.5 Gbps
 
24 GB G6 384b
Thumbnail
NVIDIA A100 SXM 80GB
 
6912
 
1410 MHz
 
3.2 Gbps
 
80 GB 5120b
Thumbnail
NVIDIA A800 SXM 80GB
 
6912
 
1410 MHz
 
3.2 Gbps
 
80 GB 5120b
Thumbnail
NVIDIA A100 PCIe 80GB
 
6912
 
1410 MHz
 
3 Gbps
 
80 GB 5120b
Thumbnail
NVIDIA A800 PCIe 80GB
 
6912
 
1410 MHz
 
3 Gbps
 
80 GB 5120b
Thumbnail
NVIDIA EGX A100 TBC TBC TBC TBC
Thumbnail
NVIDIA A100 SXM 40GB
 
6912
 
1410 MHz
 
2.4 Gbps
 
40 GB HB2 5120b
Thumbnail
NVIDIA A100 PCIe 40GB
 
6912
 
1410 MHz
 
2.4 Gbps
 
40 GB HB2 5120b
Thumbnail
NVIDIA A800 PCIe 40GB
 
6912
 
1410 MHz
 
2.4 Gbps
 
40 GB HB2 5120b
Thumbnail
NVIDIA A800 PCIe 40GB Active
 
6912
 
1410 MHz
 
2.4 Gbps
 
40 GB HB2 5120b
Thumbnail
NVIDIA A30
 
3584
 
1440 MHz
 
2.4 Gbps
 
24 GB HB2 3072b
Thumbnail
NVIDIA A2
 
1280
 
1770 MHz
 
12.5 Gbps
 
16 GB G6 128b
 ModelCoresBoost ClockMemory ClockMemory Config.
Thumbnail
NVIDIA HGX A100 (16-GPU)
 
110592
 
1410 MHz
 
2.4 GB/s
 
10240 GB HB2 5120b
Thumbnail
NVIDIA DGX A100
 
55296
 
1410 MHz
 
2.4 GB/s
 
2560 GB HB2 5120b
Thumbnail
NVIDIA HGX A100 (8-GPU)
 
55296
 
1410 MHz
 
2.4 GB/s
 
2560 GB HB2 5120b
Thumbnail
NVIDIA HGX A100 (4-GPU)
 
27648
 
1410 MHz
 
2.4 GB/s
 
640 GB HB2 5120b
Thumbnail
NVIDIA A100 SXM 80GB
 
6912
 
1410 MHz
 
3.2 GB/s
 
80 GB 5120b
Thumbnail
NVIDIA A800 SXM 80GB
 
6912
 
1410 MHz
 
3.2 GB/s
 
80 GB 5120b
Thumbnail
NVIDIA A100 PCIe 80GB
 
6912
 
1410 MHz
 
3 GB/s
 
80 GB 5120b
Thumbnail
NVIDIA A800 PCIe 80GB
 
6912
 
1410 MHz
 
3 GB/s
 
80 GB 5120b
Thumbnail
NVIDIA EGX A100 TBC TBC TBC TBC
Thumbnail
NVIDIA A100 SXM 40GB
 
6912
 
1410 MHz
 
2.4 GB/s
 
40 GB HB2 5120b
Thumbnail
NVIDIA A100 PCIe 40GB
 
6912
 
1410 MHz
 
2.4 GB/s
 
40 GB HB2 5120b
Thumbnail
NVIDIA A800 PCIe 40GB
 
6912
 
1410 MHz
 
2.4 GB/s
 
40 GB HB2 5120b
Thumbnail
NVIDIA A800 PCIe 40GB Active
 
6912
 
1410 MHz
 
2.4 GB/s
 
40 GB HB2 5120b
Thumbnail
NVIDIA CMP 170HX
 
4480
 
1410 MHz
 
2.9 GB/s
 
8 GB HB2 4096b
Thumbnail
NVIDIA A30
 
3584
 
1440 MHz
 
2.4 GB/s
 
24 GB HB2 3072b

The NVIDIA® A30 Tensor Core GPU delivers a versatile platform for mainstream enterprise workloads like AI inference, training, and high-performance computing (HPC). It combines 3rd Generation Tensor Cores with 24 GB of HBM2 memory in a dual-slot 10.5-inch PCI Express Gen4 form factor, with 165 W maximum board power. The card is passively cooled that requires system airflow to operate
within its thermal envelope.

Built on the latest NVIDIA Ampere architecture, the NVIDIA A30 brings innovations like Tensor Float 32 (TF32) and Tensor Core FP64, as well as end-to-end software stack solutions including the NVIDIA AI Enterprise suite, to ensure that mainstream AI and HPC jobs can be rapidly solved. In addition to these features, the A30 supports double precision (FP64), single precision (FP32), half precision (FP16), Brain Float 16 (BF16) and Integer (INT8) computations, unified virtual memory, and page migration engine capability. The Multi-Instance GPU (MIG) feature ensures quality of service (QoS) with secure, hardware-partitioned, right-sized GPUs across all compute workloads for a diverse set of users and maximizes the utilization of GPU resources.

The NVIDIA A30 ships with ECC enabled to protect the GPU’s memory interface and the on-board memories from detectable errors. A30’s HBM2 memory has native support for ECC with no ECC overhead, both in memory capacity and bandwidth. The card is designed to meet the requirements of NEBS Level 3 compliant servers and supports security features like secure boot and hardware root-of-trust.