NextSilicon Unveils Maverick-2 Dataflow Processor Challenging Traditional CPU/GPU Architecture
After eight years of development and $303 million in funding, NextSilicon has launched its Maverick-2 dataflow processor. The architecture reportedly dedicates most silicon to computation rather than overhead, challenging traditional CPU and GPU designs. Early deployment is expected at Sandia National Laboratory.
Revolutionary Dataflow Architecture Emerges
After eight years in development and $303 million across multiple funding rounds, NextSilicon has officially launched its Maverick-2 dataflow engine, according to company announcements. The 64-bit processor represents what analysts suggest could be a fundamental shift in computing architecture, particularly targeting high-performance computing (HPC) applications that remain heavily dependent on 64-bit floating point operations.