高性能计算发展概述.ppt

However, the Parallelizing Compilers After 30 years of intensive research only limited success in parallelism detection and program transformations instruction-level parallelism at the basic-block level can be detected parallelism in nested for-loops containing arrays with simple index expressions can be analyzed analysis techniques, such as data dependence analysis, pointer analysis, flow sensitive analysis, abstract interpretation, ... when applied across procedure boundaries often take far too long and tend to be fragile, i.e., can break down after small changes in the program instead of training compilers to recognize parallelism, people have been trained to write programs that parallelize A simple example Loop is a simple example of a code region that can benefit from parallelism Let’s look at one of the possible implementations of parallel for-loop Things to Consider in Creating a Parallelized “for-loop” Step 1 Things to Consider in Creating a Parallelized “for-loop” Step 2 Parallel Programming Complexity Enough parallelism? (Amdahl’s Law) Granularity Locality Load balance Coordination and Synchronization All of these things makes parallel programming even harder than sequential programming Parallel Compared to Sequential Programming Has different costs, different advantages Requires different, unfamiliar algorithms Must use different abstractions More complex to understand a program’s behavior More difficult to control the interactions of the program’s components Knowledge/tools/understanding more primitive References The content expressed in this chapter comes from UC Berkeley open course (/2010bootcampagenda) Carnegie Mellon University’s public course, Parallel Computer Architecture and Programming (CS 418) (/afs/cs/academic/class/15418-s11/public/lectures/) Livermore Computing Center’s training materials, Introduction to Parallel Computing (/tutorials/parallel_comp/) Parallel Programming Principle and Practice，(/courses/parallel/) 修改 * * 科学计算与模拟是推动