OBJECT-ORIENTED PROGRAMMING, SYSTEMS, LANGUAGES and APPLICATIONS
 
 
Program
 


Program (2mb PDF)

Explore
  Invited Speakers
  Onward!
  Panels
  Workshops
Discover
  Research Papers
  Student Research Comp.
  Posters
  Doctoral Symposium
  Educators' Symposium
  Wiki Symposium
  Dynamic Lang. Symp.
Understand
  Tutorials
  Essays
  Practitioner Reports
  Demonstrations
Create
  DesignFest
  Lightning Talks
  FlashBoF
  Instant Arts School Exp.
 
Other Events
 
Resort Map (364kb PDF)
 
Resort Map (JPG)

 

 
Basket
 

view, help

"Fortune Teller: Improving Garbage Collection Performance in Server Environment using Live Objects Prediction"

 

 
Page
 

Printer-friendly

 
 
  > Doctoral Symposium



Fortune Teller: Improving Garbage Collection Performance in Server Environment using Live Objects Prediction

Feng Xian, University of Nebraska-Lincoln

Currently, the most adopted criterion to invoke garbage col- lection is heap space exhaustion. In other words, garbage collection is invoked when the heap space (either the entire space or generational space) runs out. A much more dif- ficult alternative is to invoke garbage collection whenever good efficiency can be obtained. To do so, we need to know number of garbage objects that can be collected before the collection actually takes place.

In this short abstract, we introduce a predictive approach that can provide an accurate estimation of the number of dead objects at any specific point of execution. The pro- posed estimation model relies on the information obtained from partial reference counting. Our plan is to use this in- formation as a criterion to invoke garbage collection. We have conducted a preliminary study to determine the feasi- bility of this idea and found that the model is sufficiently accurate in three SPECjvm98 benchmark applications.

KEYWORDS: garbage collection, predictive model

 
.