Chapter 3 Reconfigurable Preexecution in Data Parallel Applications on Multicore Systems

摘要

The term data intensive applications or data intensive computing [1-3] is used to name a specific class of applications. Examples are web search providers, medical imaging, astronomical data analysis, weather simulation, or analyzing data collected during physical experiments. These problems all involve processing large amounts of data. The performance of such applications is usually limited by data access times rather than by pure computational power.