Processing elements (PEs) : Every processing element consists of ALU, local memory and its registers for storage of distributed data. This PEs has been interconnected via an interconnection network. All PEs get the instructions by the control unit and the dissimilar component operands are fetched by their local memory. Therefore, all PEs perform the similar function synchronously in a lock-step fashion under the control of the CU.
It might be possible that all PEs need not participate in the implementation of a vector instruction. Thus, it is required to adopt a masking scheme to control the status of every PE.
A masking vector is used to manage the status of all PEs such that only enabled PEs are allowed to participate in the implementation and others are disabled.