For almost ten years, Penguin Computing™ has been a partner of Meta (formerly known as Facebook), supporting the design and deployment of the Meta AI Research team’s high-performance computing infrastructure. In 2017, Penguin engineers helped Meta’s AI research team build their first generation of this custom-built AI infrastructure, writing close to tens of thousands lines of code to automate and optimize the infrastructure. Up until now, this infrastructure has set the bar for Meta’s researchers in terms of its performance, reliability, and productivity.
Today, Penguin Computing is proud to have provided AI optimized architecture and managed services for the newly unveiled AI Research SuperCluster (RSC) – Meta’s new cutting-edge AI supercomputer for AI research. The RSC is currently among the fastest AI supercomputers running today and is expected to be the fastest AI supercomputer in the world once it’s fully built out in mid-2022. A big part of achieving this AI milestone was in working with a number of long-time partners including Penguin Computing, NVIDIA, and Pure Storage, all of whom played a role in helping to design the first generation of Meta’s AI infrastructure back in 2017.
The development of the RSC didn’t come without challenges – namely caused by the coronavirus pandemic. Thus, RSC began as a completely “work together remotely” project that the team took from a simple shared document to a functioning cluster in about a year and a half. COVID-19 and industry-wide wafer supply constraints also brought supply chain issues that made it difficult to get everything from chips to components such as optics, GPUs, and even construction materials – all of which had to be transported in accordance with new and ever-evolving safety protocols.
Since the successful announcement of RSC, the development of Meta’s AI supercomputer is continually ongoing, and we are proud to be an integral part of the next generation of advanced AI. For example, one day AI will be able to power real-time voice translations to large groups of people, each speaking a different language, so they can seamlessly collaborate on a research project. The possibilities are truly endless, and we are helping to pave the way for the next major computing platform – the metaverse.