This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Introduction to Parallel Computing Rajendra V. Boppana CS Department UT San Antonio What is Parallel Computing Divide a task into smaller subtasks and complete each subtask concurrently with other tasks Task Divide Solve CS 6643 F11 Lec01 2 Integrate Sequential Two Types of Parallel Computing Parallel processing: Using several CPUs or computers to solve a given problem faster Processors may have to communicate with one another frequently to solve the problem Problem itself must be solved completely to obtain useful results Scientific applications Distributed computing: Using several computers to solve problem faster a problem faster Interprocessor communication cost is high and is not used frequently Problem can be decomposed into several relatively independent units of work Transaction processing Both techniques are closely related to each other CS 6643 F11 Lec01 3 A Pitfall of Parallel Computing Task Divide Solve Efficient parallel computing is challenging Dividing the problem, interprocessor communication, and integrating the solutions cause performance loss Communication requirements and...
View Full Document
- Fall '08