GPUs are currently a primary means of accelerating AI training and inference workloads, and Penguin provides a number of GPU-centered options. But it is also exploring other architectures for AI, such as GraphCore and RISC-V-based silicon. Penguin has had its GraphCore server platform ready for a year and is eagerly awaiting the first cards. Samples are expected very soon, and Pokorny expects to put the new technology in customers’ hands later this year or early next.
RISC-V is an open, free instruction set architecture based on established reduced instruction set computing principles. The company bought one of the first stand-alone RISC-V Linux machines from SiFive earlier this year and has put some of them in the hands of its government customers, who are similarly interested in forward-looking CPU architectures. Pokorny sees RISC-V as “an open CPU architecture with the potential to dramatically lower costs and potentially give users access to specialized instructions for novel new architectures that could be enabled by not having a licensing tax.”
To read more, please visit: https://www.enterprisetech.com/2018/10/19/penguin-practice-for-piecing-ai-strategies-together/.
AI stands before the HPC industry as a beacon of great expectations, yet market research repeatedly shows that AI adoption is commonly stuck in the talking phase, on the near side of a difficult chasm to cross. In response, many vendors have built integrations of their server and storage platforms with AI hardware – usually Nvidia GPUs – designed to leapfrog the initial AI implementation steps that would otherwise require the rate skills of data and computer scientists.
Penguin Computing this week announced its own strategy for helping organizations jumpstart their AI journeys, launching a new AI practice that the company says will operate as a full-service consultancy, providing guidance on system design, building custom technology solutions, and delivering professional services and support.