Abstract
A space-time mapping that describes a systolic array consists of a scheduling vector that specifies the temporal distribution and an allocation matrix that specifies the spatial distribution. The allocation matrix determines whether a variable is moving or stationary. The space-time mapping provides a complete description of the data flow for moving variables. But it fails to provide any help in handling stationary variables. This paper presents a method for the loading, recovery and access of stationary data in a systolic array. The basic idea is to add new variables, which constitute a specification for the loading, recovery and access of stationary data, to the initial program. This specification leaves open the choice of dependence vectors of the new variables and the scheduling vector. Thus, the search of schemes for handling stationary data reduces to a search of dependence vectors for the new variables and of scheduling vectors.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
