How are vector machines (e.g. the Cray-1 and its successors and imitators) similar to, and how are they different from, array processors (Flynn's SIMD category of machines)? Are they suitable for solving the same types of problems? Why or why not? And why has the vector class of machines, which once dominated the Top 500 supercomputer list, fallen out of favor in recent years?
Question
Solution Preview
This material may consist of step-by-step explanations on how to solve a problem or examples of proper writing, including the use of citations, references, bibliographies, and formatting. This material is made available for the sole purpose of studying and learning - misuse is strictly forbidden.
Vector machines, also like array processors perform the specified operation by a single machine instruction. The two classes differ mainly by the level of involved parallelism. If the vector machines are usually not parallel machines, the array processors use intensively the spatial parallelism. The difference results from the pipeline utilization; the vectors use only few pipelined execution units that overlap their computations in time....
Return to homework library