Welcome!
To use the personalized features of this site, please log in or register.
If you have forgotten your username or password, we can help.
|
 |
A Framework for Persistence-Enabled Optimization of Java Object Stores
| |
|
A Framework for Persistence-Enabled Optimization of Java Object Stores
David Whitlock6 and Antony L. Hosking6 
| (6) |
Department of Computer Sciences, Purdue University, West Lafayette, IN 47907-1398, USA |
Abstract
Aggressive optimization of programs often relies on analysis and transformation that cuts across the natural abstraction boundaries
of the source programming language, such as procedures in procedural languages, or classes in class-based object-oriented
languages like Java. Unfortunately, execution environments for languages such as Java dynamically link code into the application
as it executes, precluding cross-cutting analyses and optimizations that are too expensive to apply on-line.
Fortunately, persistent object systems usually treat the code base as an integral part of the persistent store. This code
base approximates the notion of “whole-program” that has been exploited in other optimization frameworks. This paper describes
an analysis and optimization framework for Java that operates against the persistent code base, and couples the results of
analysis and optimization with the run-time system to ensure continued correctness of the resulting code. The framework performs
extensive analysis over the code in the store, supporting optimizations that cut across class boundaries in ways that are
not safe to perform off-line on stand-alone Java classes.
Fulltext Preview (Small, Large)
 References secured to subscribers.
|
|
|
|
|
|