Skip to main navigation Skip to search Skip to main content

Using Method Lookup Caches and Register Windowing to Speed Up Dynamically-Bound Object-Oriented Applications

  • Intel
  • University of Notre Dame

Research output: Contribution to journalConference articlepeer-review

Abstract

The implementation of dynamically bound object oriented pro-gramming languages require the ability to quickly bind a logical reference to a method and to quickly allocate a context for the in-voked method. In this paper we examine how a method lookup cache (MLC) and a register windowing mechanism can speed up method binding and context allocation. We also show how the MLC can be incorporated into a contemporary pipelined datapath. A detailed register level simulation of the proposed scheme, driven by a set of fairly intensive object-oriented applications, show that a relatively small method lookup cache with only 64 or 128 entries reduces the average execution time of the applications by about 50%. With register windowing and the MLC, execution time get reduced by 76% to 87% with respect to the base machine.

Original languageEnglish
Pages (from-to)441-448
Number of pages8
JournalProceedings - Euromicro Conference on Real-Time Systems
DOIs
StatePublished - 1996
Event22nd EUROMICRO Conference on Beyond 2000: Hardware and Software Design Strategies, EUROMICRO 1996 - Prague, Czech Republic
Duration: Sep 2 1996Sep 5 1996

Keywords

  • Java
  • Smalltalk
  • dynamic binding
  • method lookup caches
  • pi-pelining
  • register windowing

Fingerprint

Dive into the research topics of 'Using Method Lookup Caches and Register Windowing to Speed Up Dynamically-Bound Object-Oriented Applications'. Together they form a unique fingerprint.

Cite this