Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Time-triggered garbage collection - Robust and adaptive real-time GC scheduling for embedded systems

Robertz, Sven LU and Henriksson, Roger LU (2003) 2003 ACM SIGPLAN conference on Language, compiler, and tool support for embedded systems 38(7). p.93-102
Abstract
The advent of Java and similar languages on the real-time system scene necessitates the development of efficient strategies for scheduling the work of a garbage collector in a non-intrusive way. We propose a scheduling strategy, time-triggered garbage collection, based on assigning the collector a deadline for when it must complete its current cycle. We show that a time-triggered CC with fixed deadline can have equal or better real-time performance than an allocation-triggered CC, which is the standard approach to real-time CC. Also, by using a deadline-based approach, the CC scheduling and, consequently, real-time performance, is independent of a complex and error-prone CC work metric. Time-triggered CC allows a more high-level view on CC... (More)
The advent of Java and similar languages on the real-time system scene necessitates the development of efficient strategies for scheduling the work of a garbage collector in a non-intrusive way. We propose a scheduling strategy, time-triggered garbage collection, based on assigning the collector a deadline for when it must complete its current cycle. We show that a time-triggered CC with fixed deadline can have equal or better real-time performance than an allocation-triggered CC, which is the standard approach to real-time CC. Also, by using a deadline-based approach, the CC scheduling and, consequently, real-time performance, is independent of a complex and error-prone CC work metric. Time-triggered CC allows a more high-level view on CC scheduling; we look at the CC cycle level rather than at the individual work increments. This makes it possible to schedule CC as any other thread. It is also suitable for making the CC auto-tuning by dynamically adjusting its deadline as necessary. We have implemented our approach in a run-time system for Java and present experimental data to support the practical feasibility of the approach. (Less)
Please use this url to cite or link to this publication:
author
and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
performance, reliability, real-time, garbage collection, auto-tuning, scheduling, embedded systems
host publication
ACM SIGPLAN Notices (Proceedings of the 2003 ACM SIGPLAN conference on Language, compiler, and tool support for embedded systems)
volume
38
issue
7
pages
93 - 102
publisher
Association for Computing Machinery (ACM)
conference name
2003 ACM SIGPLAN conference on Language, compiler, and tool support for embedded systems
conference dates
0001-01-02
external identifiers
  • wos:000185026300010
  • scopus:1442313348
ISSN
0362-1340
ISBN
1-58113-647-1
DOI
10.1145/780731.780745
language
English
LU publication?
yes
id
62ff6e12-1cbd-40bc-845e-adb7b8247293 (old id 302355)
date added to LUP
2016-04-01 15:40:59
date last changed
2022-02-20 00:36:05
@inproceedings{62ff6e12-1cbd-40bc-845e-adb7b8247293,
  abstract     = {{The advent of Java and similar languages on the real-time system scene necessitates the development of efficient strategies for scheduling the work of a garbage collector in a non-intrusive way. We propose a scheduling strategy, time-triggered garbage collection, based on assigning the collector a deadline for when it must complete its current cycle. We show that a time-triggered CC with fixed deadline can have equal or better real-time performance than an allocation-triggered CC, which is the standard approach to real-time CC. Also, by using a deadline-based approach, the CC scheduling and, consequently, real-time performance, is independent of a complex and error-prone CC work metric. Time-triggered CC allows a more high-level view on CC scheduling; we look at the CC cycle level rather than at the individual work increments. This makes it possible to schedule CC as any other thread. It is also suitable for making the CC auto-tuning by dynamically adjusting its deadline as necessary. We have implemented our approach in a run-time system for Java and present experimental data to support the practical feasibility of the approach.}},
  author       = {{Robertz, Sven and Henriksson, Roger}},
  booktitle    = {{ACM SIGPLAN Notices (Proceedings of the 2003 ACM SIGPLAN conference on Language, compiler, and tool support for embedded systems)}},
  isbn         = {{1-58113-647-1}},
  issn         = {{0362-1340}},
  keywords     = {{performance; reliability; real-time; garbage collection; auto-tuning; scheduling; embedded systems}},
  language     = {{eng}},
  number       = {{7}},
  pages        = {{93--102}},
  publisher    = {{Association for Computing Machinery (ACM)}},
  title        = {{Time-triggered garbage collection - Robust and adaptive real-time GC scheduling for embedded systems}},
  url          = {{http://dx.doi.org/10.1145/780731.780745}},
  doi          = {{10.1145/780731.780745}},
  volume       = {{38}},
  year         = {{2003}},
}