The NVIDIA A30 Tensor Core GPU, built on the advanced NVIDIA Ampere architecture, delivers unparalleled performance for enterprise workloads. Equipped with 24GB of high-bandwidth memory (HBM2) and offering up to 933 GB/s of memory bandwidth, the A30 accelerates AI training, inference, and high-performance computing (HPC) applications with exceptional efficiency and speed.
Multi-Instance GPU (MIG) TechnologyDesigned to maximise resource utilisation, the A30 supports Multi-Instance GPU (MIG) technology, enabling partitioning of a single GPU into up to four isolated instances. Each instance offers dedicated resources, making it ideal for multi-tenant environments, virtual desktop infrastructure (VDI), and concurrent AI inference tasks, all with guaranteed performance quality.
High-Performance Tensor CoresFeaturing third-generation Tensor Cores, the A30 delivers mixed-precision performance up to 20x faster than traditional FP32. With support for FP64, TF32, FP16, INT8, and INT4 precision, this GPU offers unmatched flexibility and speed for AI model training, real-time conversational AI, and complex simulations, ensuring superior outcomes for demanding HPC workloads.
Enterprise-Grade Security and IntegrationThe A30 is designed with advanced security features, including hardware root-of-trust and secure boot, ensuring data protection and compliance. With its full-height, full-length (FHFL) PCIe Gen4 form factor, it integrates seamlessly into existing server infrastructures. NVLink support allows high-speed data transfer between multiple A30 GPUs for scalable configurations and accelerated data processing.
Optimised for AI, HPC, and VirtualisationKey applications for the NVIDIA A30 include AI-driven data analytics, high-performance computing (HPC), media processing, and virtualisation. Its power-efficient 165W TDP ensures optimal performance per watt, reducing operational costs while delivering high-value results for data-intensive workloads.
Elevate your data centre capabilities with the NVIDIA A30 Tensor Core GPU, a perfect balance of performance, flexibility, and efficiency to meet the evolving demands of AI, HPC, and enterprise computing.
Part Numbers | 03GX237 |
---|---|
Brand | Lenovo |
GPU Chipset Manufacturer | NVIDIA |
RAM Bus Width | 3072 |
Render Ouput Units (ROPs) | 96 |
Texture Mapping Units (TMUs) | 224 |
RAM | 24 GB |
Cooling Component | Heatsink only |
Quick Specs: | Nvidia A30 Tensor Core GPU Data Sheet |
Power | 160W |
GPU Model | NVIDIA A30 |
GPU Memory Size | 24 GB |
GPU Memory Type | HBM2 |
Card Slot | PCI Express 4.0 x16 |
Card Form Factor | Full Height |
Power Cable Requirement | 8-Pin CPU EPS |