NVIDIA Datacenter GPU Solutions from Microway
NVIDIA GPUs are the leading acceleration platform HPC and AI. They offer overwhelming speedups compared to CPU only platforms on thousands of applications, simple directive-based programming, and the opportunity for custom code. As an NVIDIA NPN Elite partner, Microway can custom architect a bleeding-edge GPU solution for your application or code.
NVIDIA H100 GPUs

The NVIDIA H100 Tensor Core GPU powered by the NVIDIA Hopper™ architecture provides the utmost in GPU acceleration for your deployment and groundbreaking features
- A dramatic leap in performance for HPC
Up to 34 TFLOPS FP64 double-precision floating-point performance (67 TFLOPS via FP64 Tensor Cores) - Unprecedented performance for Deep Learning
Speedups of up to 9X for AI training and up to 30X for inference with Transformers, approaching 2 PFLOPS/4PFLOPS performance - 2nd Gen Multi Instance GPU (MIG)
Provides up to 7X the secure tenants for more, fully isolated isolated applications - Fastest GPU memory
80GB of HBM3 operating at up to 3.35TB/s - Class-leading connectivity
PCI-E Gen 5 and for up to 128GB/sec of transfer BW to DPUs or InfiniBand and 4th-generation NVLink for up to 900GB/sec of GPU-GPU communication
NVIDIA A100 GPUs
NVIDIA A100 “Ampere” GPUs provide advanced GPU acceleration for your deployment and offer advanced features
- Multi Instance GPU (MIG)
Allows each A100 GPU to run seven separate & isolated applications or user sessions - Strong HPC performance
Up to 9.7 TFLOPS FP64 double-precision floating-point performance (19.5 TFLOPS via FP64 Tensor Cores) - Excellent Deep Learning throughput
Speedups of 3x~20x for neural network training and 7x~20x for inference (vs Tesla V100) and new TF32 instructions - Large, Fast GPU memory
80GB of high-bandwidth memory operating at up to 2TB/s - Faster connectivity
3rd-generation NVLink provides 10x~20x faster transfers than PCI-Express

NVIDIA A30 GPUs

NVIDIA A30 “Ampere” GPUs offer versatile compute acceleration for mainstream enterprise GPU servers
- Amazing price-performance for HPC compute
Up to 5.2 TFLOPS FP64 double-precision floating-point performance (10.3 TFLOPS via FP64 Tensor Cores) - Strong AI Training & Inference Performance
approximately ~50% of the FP16 Tensor FLOPS of an NVIDIA A100 and support for TF32 instructions - Large and fast GPU memory spaces
24GB of high-bandwidth memory with 933GB/s of memory bandwidth - Multi Instance GPU (MIG)
Allows each A30 GPU to run four separate & isolated applications or user sessions - Fast connectivity
PCI-Express Gen 4.0 interface to host and 200GB/sec 3rd Gen NVLink Interface to neighboring GPUs
Why NVIDIA Datacenter GPUs?
NVIDIA Datacenter GPUs (formerly Tesla GPUs) have unique capabilities not present in consumer GPUs and are the ideal choice for professionals deploying clusters, servers, or workstations. Unique features to NVIDIA’s datacenter GPUs include:
Full NVLink Capability, Up to 900GB/sec
Only NVIDIA Datacenter GPUs deploy the most robust implementation of NVIDIA NVLink for the highest bandwidth data transfers. At up to 900GB/sec per GPU, your data moves freely throughout the system and nearly 14X the rate of PCI-E x16 4.0 GPUs.
Unique Instructions for AI Training, AI Inference, & HPC
Datacenter GPUs support the latest TF32, BFLOAT16, FP64 Tensor Core, Int8, FP8 instructions that dramatically improve application performance.
Unmatched Memory Capacity, up to 80GB per GPU
Support your largest datasets with up to 80GB of GPU memory, far greater capacity than available on consumer offerings.
Full GPU Direct Capability
Only datacenter GPUs support the complete array of GPU Direct P2P, RDMA, and Storage features. These critical functions remove unnecessary copies and dramatically improve data flow.
Explosive Memory Bandwidth up to 3TB/s and ECC
NVIDIA Datacenter GPUs uniquely feature HBM2 and HBM3 GPU memory with up to 3TB/sec of bandwidth and full ECC protection.
Superior Monitoring & Management
Full GPU integration with the host system’s monitoring and management capabilities such as IPMI. Administrators can manage datacenter GPUs with their widely-used cluster/grid management tools.