In this video, Yaniv Sapir from Adapteva demonstrates the the Epiphany-IV 28nm multicore silicon performance. The video shows how the Epiphany processor beats an x86 processor while executing a parallel matrix multiplication demo written entirely in ANSI-C.
The Epiphany microprocessor architecture is a supremely scalable shared memory architecture, featuring up to 4,096 processors on a single chip, connected through a high-bandwidth on-chip network. Each processor node represents a fully-featured floating point RISC processor built from scratch for multicore processing, a high bandwidth local memory system, and an extensive set of built in hardware features for multicore communication. The resulting performance boost is coupled with Adapteva’s low power design and standard C programming model, bringing an unprecedented level of real-time processing to performance and power constrained mobile devices like smartphones and tablet computers, as well as improving performance levels for an array of other parallel computing platforms.
Read the Full Story over at GPU Science.