Displaying items by tag: Nvidia A100

Nvidia DGX Cloud gives enterprises immediate access to the infrastructure and software needed to train advanced models for generative AI and other applications.

Nvidia's next-generation H100 Tensor Core GPUs and Quantum-2 InfiniBand are now widely available, in Microsoft Azure and more than 50 partner systems from the company's partners including Asus, Atos, Dell Technologies, Gigabyte, HPE, Lenovo, and Supermicro.

Nvidia H100 GPUs set new records in all eight of the MLPerf Training benchmarks, while the A100 came top in the latest round of MLPerf HPC benchmarks.

MLPerf Inference provides an indication of how well systems perform at inferencing tasks such as image classification and speech recognition. Nvidia's A100 achieved the highest performance per accelerator in the Data Center Offline and Server tests.

The Permutter supercomputer at the US National Energy Research Scientific Computing Center (NERSC) will deliver approximately four exaflops of AI performance, according to GPU supplier Nvidia, making it the world's fastest on this measure.

The latest version of Nvidia's virtual GPU (vGPU) technology makes GPU-accelerated virtual machines available from the data centre or the cloud.

HPC and AI vendor Nvidia has introduced an upgraded GPU, a new workgroup server, and a next-generation networking technology.

GPU vendor Nvidia set records in the latest round of MLPerf Inference benchmarking.

Oracle claims to be the first major cloud provider to make Nvidia A100 Tensor Core GPU on bare metal instances.

GPU maker Nvidia has dominated the commercially-available categories in the MLPerf Training v0.7 benchmark results.

Eight of the top 10 supercomputers on the latest Top500 list use Nvidia GPUs, Nvidia Mellanox InifiniBand networking, or both. Nvidia used the occasion to make several other announcements.

GPU maker Nvidia has made a slew of announcements at its GTC 2020 event, which was held online rather than in its usual conference format.