Scan code: LN116071 Manufacturer code: TCSA30M-PB

PNY NVIDIA A30 24GB Passive Ampere Graphics Card

PNY NVIDIA A30 24GB HBM2 with ECC Passive Graphics Card, 3804 CUDA, 5.2 TFLOPS DP, 10.3 TFLOPS SP, 165 TFLOPS HP

Thumbnail 1 : PNY NVIDIA A30 24GB Passive Ampere Graphics Card Thumbnail 2 : PNY NVIDIA A30 24GB Passive Ampere Graphics Card Thumbnail 3 : PNY NVIDIA A30 24GB Passive Ampere Graphics Card
Scan code: LN116071 Manufacturer code: TCSA30M-PB
£4,679.99
Order now and get it tomorrow, Mon 2nd Aug
With a 10% deposit from just £116.89 / month (APR 15.9%)
Delivery Options
  • By DPD On 3rd Aug to your specified address. | £11.50 Receive SMS with one-hour delivery window Weekend, timed and European delivery options are available at checkout

    Have your parcel delivered by DPD to your specified address. Receive SMS with one-hour delivery window

    Weekend, timed and European delivery options are available at checkout
  • Collect instore Collect from our Bolton store, BL6 6PE | Free

    Order online, collect from our Bolton store (25-28 Enterprise Park, Middlebrook, Horwich, Bolton, BL6 6PE)

Services
  • 48HR REPLACEMENT If you need to return this item, your replacement will be dispatched within 2 working days of your product arriving back at Scan. More info

PNY NVIDIA A30 Tensor Core GPU with Ampere Architecture

A30 GPU

AI Inference and Mainstream Compute for Every Enterprise NVIDIA A30 Tensor Core GPU is the most versatile mainstream compute GPU for AI inference and mainstream enterprise workloads. Powered by NVIDIA Ampere architecture Tensor Core technology, it supports a broad range of math precisions, providing a single accelerator to speed up every workload. Built for AI inference at scale, the same compute resource can rapidly re-train AI models with TF32, as well as accelerate high-performance computing (HPC) applications using FP64 Tensor Cores. Multi-Instance GPU (MIG) and FP64 Tensor Cores combine with fast 933 gigabytes per second (GB/s) of memory bandwidth in a low 165W power envelope, all running on a PCIe card optimal for mainstream servers.

24GBHBM2 GPU Memory

PCI Express 4.0 x16System interface

3,804CUDA Cores

933 GB/sMemory Bandwidth

Groundbreaking Innovative Technology

down

NVIDIA AMPERE ARCHITECTUREWhether using MIG to partition an A30 GPU into smaller instances, or NVLink to connect multiple GPUs to accelerate large-scale workloads, A30 can readily handle different-sized acceleration needs, from the smallest job to the biggest multi-node workload. A30's versatility means IT managers can maximize the utility of every GPU in their data center around the clock.

up

THIRD-GENERATION TENSOR CORESNVIDIA A30 delivers 165 teraFLOPS (TFLOPS) of TF32 deep learning performance. That’s 20X more AI training throughput and over 5X more inference performance compared to NVIDIA T4 Tensor Core GPU. For HPC, A30 delivers 10.3 TFLOPS of performance, nearly 30 percent more than NVIDIA V100 Tensor Core GPU.

up

NEXT-GENERATION NVLINKNVIDIA NVLink in A30 delivers 2X higher throughput compared to the previous generation. Two A30 PCIe GPUs can be connected via an NVLink Bridge to deliver 330 TFLOPS of deep learning performance.

up

MULTI-INSTANCE GPU (MIG)An A30 GPU can be partitioned into as many as four GPU instances, fully isolated at the hardware level with their own high-bandwidth memory, cache, and compute cores. MIG gives developers access to breakthrough acceleration for all their applications. IT administrators can offer right-sized GPU acceleration for every job, optimizing utilization and expanding access to every user and application.

up

HBM2With 24 gigabytes (GB) of high-bandwidth memory (HBM2), with a bandwidth of 933 gigabytes per second (GB/s), researchers can rapidly solve double-precision calculations. HPC applications can also leverage TF32 to achieve higher throughput for single-precision, dense matrix-multiply operations.

up

STRUCTURAL SPARSITYAI networks have millions to billions of parameters. Not all of these parameters are needed for accurate predictions, and some can be converted to zeros, making the models “sparse” without compromising accuracy. Tensor Cores in A30 can provide up to 2X higher performance for sparse models. While the sparsity feature more readily benefits AI inference, it can also improve the performance of model training.

How Fast is the A30?

Performance

Up to 1.1X higher throughput than V100 and 8X higher than T4 NVIDIA A30 features FP64 NVIDIA Ampere architecture Tensor Cores that deliver the biggest leap in HPC performance since the introduction of GPUs. Combined with 24 gigabytes (GB) of GPU memory with a bandwidth of 933 gigabytes per second (GB/s), researchers can rapidly solve double-precision calculations. HPC applications can also leverage TF32 to achieve higher throughput for single-precision, dense matrix-multiply operations.

High-Performance Data Analytics

Data analytics

analyze, visualize, and turn massive datasets into insights Accelerated servers with A30 provide the needed compute power – along with large HBM2 memory, 933GB/sec of memory bandwidth, and scalability with NVLink – to tackle these workloads. Combined with NVIDIA InfiniBand, NVIDIA Magnum IO and the RAPIDS™ suite of open-source libraries, including the RAPIDS Accelerator for Apache Spark, the NVIDIA data center platform accelerates these huge workloads at unprecedented levels of performance and efficiency.

Product Specifications
CUDA Cores 3804
NVIDIA RT Cores 72
GPU Memory Size 24 GB GDDR6 with ECC
Memory Bus Width 3072-bit
Memory Bandwidth 933 GB/s
Peak Double Performance 5.2 TFLOPS (GPU Boost Clocks)
Peak Single Performance 10,3 TFLOPS (GPU Boost Clocks)
Peak Half Precision 165 TFLOPS (GPU Boost Clocks)
System Interface PCI Express 4.0 x16
Max Power Consumption 165 W
Power Connectors 1x 8-pin CPU Power Connector
Weight (w/o extender) 1024g
Thermal Solution Passive Heatsink
Form Factor 111,15 mm (H) x 267,7 mm (L) Dual Slot, Full Height
Warranty 3 Years
Specifications
Key Specifications
Graphics Chipset NVIDIA A30
Edition PNY A30
Manufacturing process 8 nm
Microarchitecture Ampere
GPU Name  
Cores & Clocks
NVIDIA CUDA Cores 3804
Core Clock  
Boost Clock  
Video Memory (VRAM)
Memory Size 24 GB
Memory Type HBM2
Memory Clock  
Memory Bit Rate 3072 Bit
Memory Bandwidth 933 GB/s
ECC Technology Yes
Cooling
Cooling Passive
I/O & Connectivity
Interface PCIe 4.0 (x16)
Interface Bandwidth  
Graphics Output  
Visuals
Multi GPU Support  
NVLink Support N/A
Total NVLink Bandwidth N/A
Microsoft DirectX Support  
HDCP Ready No
Multi Monitor Support  
Concurent Users N/A
H.264 1080p30 Streams N/A
Maximum Digital Resolution  
Maximum VGA Resolution  
Supported Graphics APIs  
Compute Performance
Supported Compute APIs  
Single Precision (FP32) Processing Yes
Single Precision (FP32) Performance 10.3 teraFLOPS
Double Precision (FP64) Processing Yes
Double Precision (FP64) Performance 5.2 teraFLOPS
Tensor Processing No
Deep Learning (Tensor) Performance  
NVIDIA Tensor Cores  
Integer Operations (INT8)  
Ray-Tracing
NVIDIA RT Cores  
RTX-OPS  
Rays Cast  
Power & Thermals
Graphics Card Power Connectors 1 x 8-pin PCIe
Board Power 165 W
Minimum Recommended PSU  
Maximum GPU Temperature  
Physical
Form Factor Dual Slot
Low Profile Compatible No
Low Profile Support  
Dimensions 267.7 x 111.15 mm
Package Type  
Additional Information
Scan Code LN116071
Model Number TCSA30M-PB
GTIN 3536403384135
Warranty

Please note your statutory rights are not affected.

For further information regarding Scan's warranty procedure please see our terms and conditions

Warranty information has not yet been updated for this product. For further information please call 0871 472 4747 or 01204 474747.

PNY NVIDIA A30 24GB Passive Ampere Graphics Card
PNY NVIDIA A30 24GB Passive Ampere Graphics Card