A Roadmap on Very Long Instruction Word CPUs Family
Very Long Instruction Word (VLIW) refers to a processor architecture designed to take advantage of instruction level parallelism. Whereas, conventional processors mostly allow programs that specify instructions to be executed at the same time (i.e. in parallel). This type of processor architecture is intended to allow higher performance without the inherent complexity of some other approaches. A processor that executes every instruction one after the other (i.e. a non-pipelined scalar architecture) use processor resources inefficient, potentially leading to poor performance.