Filters








53 Hits in 2.0 sec

Pretenuring for Java

Stephen M. Blackburn, Sharad Singhai, Matthew Hertz, Kathryn S. McKinely, J. Eliot B. Moss
2001 SIGPLAN notices  
Pretenuring can reduce copying costs in garbage collectors by allocating long-lived objects into regions that the garbage collector will rarely, if ever, collect.  ...  noticeably better. (4) Our same advice improves the performance of generational and Older First collection, illustrating that it is collector neutral.  ...  We also thank John Cavazos, Asjad Khan, and Narendran Sachindran for their contributions to various incarnations of this work, and our anonymous reviewers for their helpful comments.  ... 
doi:10.1145/504311.504307 fatcat:3l6prlv7ejbqlj6l2r4o2y24ne

Towards intelligent analysis techniques for object pretenuring

Jeremy Singer, Gavin Brown, Mikel Luján, Ian Watson
2007 Proceedings of the 5th international symposium on Principles and practice of programming in Java - PPPJ '07  
It is a key optimization for generational garbage collection systems, which are standard in most high performance Java virtual machines.  ...  A study of garbage collection traces from four standard Java benchmark programs shows that there is high dependence on some of these factors such as allocation site and object type.  ...  INTRODUCTION Garbage collection (GC) is the process of automatic management of dynamically allocated memory [13] .  ... 
doi:10.1145/1294325.1294353 dblp:conf/pppj/SingerBLW07 fatcat:yg7bjtimrnh2pjfz22uti5y2ru

Memento mori: dynamic allocation-site-based optimizations

Daniel Clifford, Hannes Payer, Michael Stanton, Ben L. Titzer
2015 SIGPLAN notices  
This paper presents a general approach for gathering temporal allocation site feedback that tackles both the general problem of object lifetime estimation and improves optimization of these problematic  ...  Unlike previous work on pretenuring, our system utilizes allocation mementos to achieve fully dynamic allocation-sitebased pretenuring in a production system.  ...  For example, instances of this idea are dynamically adjusting promotion thresholds [30, 35] , variable-sized young generations [2] , and multiple generations with different garbage collection strategies  ... 
doi:10.1145/2887746.2754181 fatcat:7o3cbwpjyfgr3b7mpfvzom3ebe

Memento mori: dynamic allocation-site-based optimizations

Daniel Clifford, Hannes Payer, Michael Stanton, Ben L. Titzer
2015 Proceedings of the 2015 ACM SIGPLAN International Symposium on Memory Management - ISMM 2015  
This paper presents a general approach for gathering temporal allocation site feedback that tackles both the general problem of object lifetime estimation and improves optimization of these problematic  ...  Unlike previous work on pretenuring, our system utilizes allocation mementos to achieve fully dynamic allocation-sitebased pretenuring in a production system.  ...  We delivered our optimizations in the V8 production virtual machine which demanded important controls like a pretenuring recovery mechanism and comprehensiveness over a wide range of dynamic behavior.  ... 
doi:10.1145/2754169.2754181 dblp:conf/iwmm/CliffordPST15 fatcat:4k646jgtdjemrhkh5jteohfoey

Profile-based pretenuring

Stephen M. Blackburn, Matthew Hertz, Kathryn S. Mckinley, J. Eliot B. Moss, Ting Yang
2007 ACM Transactions on Programming Languages and Systems  
Pretenuring can reduce copying costs in garbage collectors by allocating long-lived objects into regions that the garbage collector will rarely, if ever, collect.  ...  We extend previous work on pretenuring as follows. (1) We produce pretenuring advice that is neutral with respect to the garbage collector algorithm and configuration.  ...  Watson Research Center who helped facilitate this research, IBM Research for making Jikes RVM widely available, and the broader Jikes RVM community for their contributions to the research platform.  ... 
doi:10.1145/1180475.1180477 fatcat:l52csauynnh5fehf7dvloy4byy

Dynamic object sampling for pretenuring

Maria Jump, Stephen M. Blackburn, Kathryn S. McKinley
2004 Proceedings of the 4th international symposium on Memory management - ISMM '04  
Many state-of-the-art garbage collectors are generational, collecting the young nursery objects more frequently than old objects.  ...  For example, pretenuring decreases nursery collection work by allocating new, but long-lived, objects directly into the mature space.  ...  We report total time, garbage collection time, and mutator time results for dynamic pretenuring for these configurations.  ... 
doi:10.1145/1029873.1029892 dblp:conf/iwmm/JumpBM04 fatcat:kzb6ngkp3fdbjfz3d4jtqpmyhe

Generational stack collection and profile-driven pretenuring

Perry Cheng, Robert Harper, Peter Lee
1998 SIGPLAN notices  
1 Introduction This paper presents two techniques for improving garbage collection performance: generational stack collection and profile-driven pretenuring.  ...  This technique, called generational stack collection, is like generational garbage collection in that old stack frames are "tenured" to reduce processing frequency.  ...  Acknowledgements Chris Stone deserves thanks for general discussions about garbage collection, advice on benchmarking, and, as usual, a thorough review of this paper.  ... 
doi:10.1145/277652.277718 fatcat:ycq3ofqvwnauvmnvc3fxgh73bi

Generational stack collection and profile-driven pretenuring

Perry Cheng, Robert Harper, Peter Lee
1998 Proceedings of the ACM SIGPLAN 1998 conference on Programming language design and implementation - PLDI '98  
1 Introduction This paper presents two techniques for improving garbage collection performance: generational stack collection and profile-driven pretenuring.  ...  This technique, called generational stack collection, is like generational garbage collection in that old stack frames are "tenured" to reduce processing frequency.  ...  Acknowledgements Chris Stone deserves thanks for general discussions about garbage collection, advice on benchmarking, and, as usual, a thorough review of this paper.  ... 
doi:10.1145/277650.277718 dblp:conf/pldi/ChengHL98 fatcat:4kjsovyrdfecdp3htqc33gw2c4

Decrypting the Java gene pool

Sebastien Marion, Richard Jones, Chris Ryder
2007 Proceedings of the 6th international symposium on Memory management - ISMM '07  
Pretenuring long-lived and immortal objects into infrequently or never collected regions reduces garbage collection costs significantly.  ...  We obtain performance gains between 6-77% in GC time against a generational copying collector for several SPEC jvm98 programs.  ...  We thank Yossi Gil and Itay Maman for providing their pattern analysis tool, IBM Research for making Jikes RVM available, the MMTk Core Team for their GC framework, and the DaCapo group for their benchmark  ... 
doi:10.1145/1296907.1296918 dblp:conf/iwmm/MarionJR07 fatcat:ublor3noynhhpgk6l5laepyzj4

On the prediction of Java object lifetimes

H. Inoue, D. Stefanovic, S. Forrest
2006 IEEE transactions on computers  
For the SPEC benchmark programs, the number of dynamically allocated objects whose call sites have accurate predictors ranges from 0.2 percent to 61 percent.  ...  This method could potentially improve the performance of garbage collectors.  ...  ACKNOWLEDGMENTS The authors are grateful to Matthew Hertz for creating and providing to them the Merlin algorithm implementation.  ... 
doi:10.1109/tc.2006.107 fatcat:e4rd7pnxhjhfjnuvimcefvmm4y

Dynamic adaptive pre-tenuring

Timothy L. Harris
2001 SIGPLAN notices  
Pretenuring long-lived objects reduces the number of times that they are scanned or copied during garbage collection.  ...  In a generational garbage collector, a pre-tenured object is one that is allocated directly in the old generation.  ...  Dynamic pre-tenuring has a marked effect on young generation garbage collection. The number of collections is reduced by 11%. The total time spent in young generation collection is reduced by 9%.  ... 
doi:10.1145/362426.362476 fatcat:ljza5wmlhrehtjc2jyvvjuykk4

Dynamic adaptive pre-tenuring

Timothy L. Harris
2000 Proceedings of the second international symposium on Memory management - ISMM '00  
Pretenuring long-lived objects reduces the number of times that they are scanned or copied during garbage collection.  ...  In a generational garbage collector, a pre-tenured object is one that is allocated directly in the old generation.  ...  Dynamic pre-tenuring has a marked effect on young generation garbage collection. The number of collections is reduced by 11%. The total time spent in young generation collection is reduced by 9%.  ... 
doi:10.1145/362422.362476 dblp:conf/iwmm/Harris00 fatcat:nvyourg5tvb3bixzim4mrwfgiq

Finding your cronies

Samuel Z. Guyer, Kathryn S. McKinley
2004 SIGPLAN notices  
Experimental results for SPEC Java benchmarks using Jikes RVM show colocation can reduce garbage collection time by 50% to 75%, and total performance by up to 10%.  ...  This paper introduces dynamic object colocation, an optimization to reduce copying costs in generational and other incremental garbage collectors by allocating connected objects together in the same space  ...  DYNAMIC OBJECT COLOCATION The goal of dynamic object colocation is to allocate connected objects in the same garbage collection space.  ... 
doi:10.1145/1035292.1028996 fatcat:v7jtbhdclnfgbcgnjdyqtv3mjm

Finding your cronies

Samuel Z. Guyer, Kathryn S. McKinley
2004 Proceedings of the 19th annual ACM SIGPLAN Conference on Object-oriented programming, systems, languages, and applications - OOPSLA '04  
Experimental results for SPEC Java benchmarks using Jikes RVM show colocation can reduce garbage collection time by 50% to 75%, and total performance by up to 10%.  ...  This paper introduces dynamic object colocation, an optimization to reduce copying costs in generational and other incremental garbage collectors by allocating connected objects together in the same space  ...  DYNAMIC OBJECT COLOCATION The goal of dynamic object colocation is to allocate connected objects in the same garbage collection space.  ... 
doi:10.1145/1028976.1028996 dblp:conf/oopsla/GuyerM04 fatcat:ytftmpwvrvbtflasdtezrafcru

Remote Objects: The Next Garbage Collection Challenge

Witawas Srisa-an, Mulyadi Oey
2005 Journal of Object Technology  
Third, we suggest various optimization techniques to improve the garbage collection efficiency in distributed objects environments.  ...  We find that the garbage collection efficiency degrades significantly when the heap is interspersed with both local and remote objects.  ...  Sebastian Elbaum for offering many valuable suggestions. This project was partially supported by the National Science Foundation under grant CNS-0411043 and Univ. of Nebraska Layman's Award.  ... 
doi:10.5381/jot.2005.4.4.a4 fatcat:shotcrfrpfbhphgcer6jlyudvi
« Previous Showing results 1 — 15 out of 53 results