Every cycle each core gets the same instruction. Based on dynamic data at runtime, each core can interpret that instruction differently.
The broadcast bus transmits loop invariant data, such as weights and constants, to all Vortex Cores at once.
Large local memories offer the developer space for large data structures. Multiple ports allow for simultaneous reading and writing to keep the vortex core array busy.
Edge load-store units are static and completely software-controlled, allowing for deterministic kernel runtimes. Each edge has a load-store unit, unlocking novel software API possibilities such as native data rotations and data remapping.
driven by code
configurable multimode multiply accumulate
general-purpose arithmetic logic unit
scalable single-cycle local memory
single-cycle data sharing
the right data.
right on time.
scalable to meet the needs of any workload