Today Groq announced that the company’s tensor processing hardware is now available on the Nimbix Cloud for select partners. The Groq processor is designed specifically for the performance requirements of computer vision, machine learning and other AI-related workloads, and is the first architecture in the world capable of 1 PetaOp/s performance on a single chip. Nimbix is the leading HPC cloud platform provider and offers purpose-built cloud computing for the world’s most demanding workloads.
Our partners are excited to have the opportunity to access Groq hardware in the Nimbix Cloud,” said Jonathan Ross, Groq’s co-founder and CEO, “and I’m thrilled to give our customers access to Nimbix’s high performance infrastructure and award winning support.”
This partnership eliminates the complications of setting up and maintaining on-site machine learning infrastructure, allowing customers instead to purchase ML acceleration as an on-demand service. They can take advantage of opportunities that derive from artificial intelligence and cognitive computing without needing to invest in hardware optimized for machine learning.
Inspired by a software-first mindset, Groq’s TSP architecture provides a new paradigm for achieving both compute flexibility and massive parallelism without synchronization overhead of traditional GPU and CPU architectures. The Groq architecture is many multiples faster than anything else available for inference, in terms of both low latency and inferences per second.
Nimbix is dedicated to offering innovative, leading-edge solutions on the Nimbix Cloud and our JARVICE platform, and this is one reason that we are proud to be working with Groq,” said Steve Hebert, Nimbix CEO. “Groq’s simplified processing architecture is unique, providing unprecedented, deterministic performance for compute intensive workloads, and is an exciting addition to our cloud-based AI and Deep Learning platform.”
