The NoC interconnect is highly configurable and has been designed to meet the demanding workload and time-to-market requirements of the next generation of high-performance, energy-efficient computing for AI, HPC, and automotive applications.
Tenstorrent designs AI graph processors, high-performance RISC-V CPUs and configurable chiplets that combine hardware and software to deliver scalable, energy-efficient computing for AI and machine learning applications for a broad range of end markets, from training to inference computing.
“We continue to leverage Arteris’ network-on-chip IP products in our designs as we drive the next wave of advancements in AI computing," said David Bennett, CCO of Tenstorrent. “Arteris is a proven technology partner - their FlexNoC IP provides superior performance for our next-generation AI compute."
"The demands of AI continue to push the envelope, but our silicon-proven IP technology was built to meet the demands of complex workloads and chiplet architectures," said K. Charles Janac, president and CEO of Arteris.
FlexNoC non-coherent network-on-chip IP is highly configurable, ensuring high-performance data movement across CPU workloads. With its advanced features such as congestion management, quality of service (QoS), and error detection and correction, FlexNoC provides low-latency, power-efficient communication between CPU clusters, delivering improved levels of performance and productivity.