Nvidia doesn’t have all the inside pieces to build its massive AI computing empire, so it’s recruiting software and hardware partners to scale its so-called AI factories in the cloud.
The chipmaker’s most recent partnership is with Rescale, which provides middleware for organizing high-performance computing workloads on public and hybrid clouds.
Rescale is adopting Nvidia’s AI technology for what the middleware provider has called the world’s first computing recommendation engine, which automates the selection of computing resources available in the cloud for high-performance applications and simulations, said Edward Hsu, chief product officer at Rescale.
“We have more than 1,000 applications and releases from leading simulation software providers, and we work with all major cloud providers and specialty architecture providers to improve the performance of computational science and engineering workloads,” Hsu said.
Rescale will also offer customers Nvidia’s AI Enterprise software platform, which includes pre-programmed AI models such as Modulus, which are used for scientific and simulation applications. Rescale can integrate enterprise AI with scientific or engineering applications from companies including ANSYS, Siemens, and others for deployment in the cloud.
Hsu said companies are starting to use more computational-intensive approaches to science and engineering discovery, and the Computing Recommendation Engine is helping customers decide what type of computing they need to use.
Some of the considerations regarding the computing requirements of customers include the type of architecture, geographies, data sovereignty, compliance, or regions where more GPU resources are available for applications such as video and artificial intelligence.
“It gets very complicated and what we’ve done is take all of our years taking this information from metadata generated from running computing functions, implementing a set of proprietary recycling programs and standards [and] TensorFlow. The work on the Nvidia GPU is already developing this computing recommendation engine,” Hsu said.
Major cloud providers already offer virtualization and orchestration technologies. Rescale already collaborates with all major cloud service providers, and is adding an account recommendation engine on top of that for customers to improve the performance of their workloads on these architectures.
More companies are taking a hybrid cloud approach to meet growing computing needs. Rescale uses Nvidia’s AI Enterprise middleware to match engineering applications with hardware resources across the cloud.
Hsu described many parts of the Nvidia partnership as an integrated offering. For example, if a customer is building a turbocharger for cars, the Rescale platform can make recommendations for change parameters and calculate resources to use to get the fastest and best results.
Dion Harris, head of data center product marketing at Nvidia, said during a newspaper directive.
Nvidia already dominates the scientific computing space, but Rescale provides an indirect way for HPC customers to adopt AI Enterprise software to train or use AI models.
Nvidia is looking to the future as it ties its hardware and software into a black box. Nvidia is designing its AI software to run more efficiently on its GPUs. Nvidia’s software suite is based on the CUDA parallel programming framework, and the company is ramping up the deployment of its latest GPUs at Huber in the cloud.
GPU Maker Last month I partnered with Oracle To bring AI Enterprise software and thousands of GPUs to Oracle’s cloud infrastructure. The company’s GPUs are also widely adopted by Meta to run AI applications.
Nvidia hardware also works with open source parallel programming frameworks such as OpenCL, operated by The Khronos Group. Intel’s OpenAPI Parallel Programming Framework contains a tool called SYCL that outputs special CUDA code to run on accelerators from Intel and other chip makers.