NVIDIA has introduced an end-to-end hyperscale data center platform that lets companies accelerate machine learning workloads. It is comprised of two accelerators. One allows researchers design new neural networks and increase the number of applications powered with artificial intelligence. The second is a low-power accelerator designed to deploy networks across the data center.
This new offering uses the Tesla Accelerated Computing platform and includes a suite of libraries that are GPU (graphics processing unit) accelerated. New additions to the platform are:
- NVIDIA Tesla M40 GPU: An accelerator designed for training deep neural networks.
- NVIDIA Tesla M4 GPU: A low-power accelerator for machine learning interference.
- NVIDIA Hyperscale Suite: Software optimized for machine learning and video processing.
The M40 GPU, according to the company, allows data scientists to save days or weeks of time training neural networks. It is optimized for machine learning, built for 24/7 reliability and scale out performance.
The NVIDIA Tesla M40 GPU provides scalability. Image courtesy of NVIDIA.
For lower-power applications, the M4 GPU offers a smaller form factor, higher throughput and a user-selectable power profile to consume 50-75 watts.
The NVIDIA Tesla M4 GPU offers low-power usage for neural network applications. Image courtesy of NVIDIA.
The Hyperscale Suite has tools for developers and data center managers, and has cuDNN, FFmpeg multimedia software, Image Compute Engine and a GPU REST Engine.
“The artificial intelligence race is on,” said Jen-Hsun Huang, co-founder and CEO of NVIDIA. “Machine learning is unquestionably one of the most important developments in computing today, on the scale of the PC, the internet and cloud computing. Industries ranging from consumer cloud services, automotive and health care are being revolutionized as we speak. “Machine learning is the grand computational challenge of our generation. We created the Tesla Hyperscale Accelerator line to give machine learning a 10X boost. The time and cost savings to data centers will be significant.”
For more information, visit NVIDIA.
Sources: Press materials received from the company and additional information gleaned from the company’s website.