Abstract
This paper examines the role of partitioning in the design of distributed data processing (DDP) systems for high technology, data driven, real time applications (e.g., Ballistic Missile Defense - BMD). Partitioning processes are contrasted with allocation processes. An overview of the DDP design process is presented, with emphasis on the virtual design phase. Contemporary partitioning strategies and concepts are reviewed, and related to system design goals. Goal-oriented partitioning considerations are discussed for three typical objectives: growth, reliability, and performance. A candidate sequence of partitioning steps to support DDP virtual design analysis is presented.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have