An algorithm for parallel incremental compaction

Ben-Yitzhak, Ori; Goft, Irit; Kolodner, Elliot K.; Kuiper, Kean; Leikehman, Victor

doi:10.1145/773039.512442

Cited by 8 publications

(12 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The space overhead depends on 1) how much memory is allocated during a compaction cycle and 2) where and how large an evacuation area is chosen. In this measurement, since the target pause time is 100 microseconds, which is much shorter than in other researches [5,9,4,3], much memory is required during a compaction cycle. This is one of the reasons why the space over head is larger.…”

Section: Space Overheadmentioning

confidence: 95%

“…[5] proposed an algorithm for parallel incremental compaction, by which we were inspired to develop our compaction. Their compaction algorithm first chooses the area to evacuate, which is called evacuation area, prior to the mark phase of the incremental mark-sweep GC.…”

Section: Related Workmentioning

confidence: 99%

“…Most incremental and/or real-time compactors [5,9,4, 3] achieved a pause time of a few milliseconds. On the other hand, our compactor achieved a pause time of about 100 microseconds, which is comparable to that caused by the snapshot collector.…”

Section: Pause Timementioning

confidence: 99%

“…during compaction. Although there are several incremental compaction algorithms [5,13,11,9], some of them do not reduce the maximum pause time to a level comparable to incremental GC algorithms and the others require readbarriers. Read-barriers involve a large overhead in a situation where no dedicated hardware such as a page protection mechanism is available.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Replication-Based Incremental Compaction

Ugawa

Yasugi

Yuasa

2008

2008 11th IEEE International Symposium on Object and Component-Oriented Real-Time Distributed Computing (ISORC)

View full text Add to dashboard Cite

We propose an incremental compaction algorithm. Our compactor selects a continuous area of the heap and evacuates it by incrementally copying all objects in the area to the rest of the heap. After all objects have been copied, our compactor incrementally updates pointers pointing into the evacuated area. During these processes, each original object and its copy are kept consistent.We implemented the compactor together with a snapshot garbage collector in the KVM. Our measurements show that (1) the largest free chunk is almost always more than 20% as large as the entire heap when the heap is twice as large as the maximum amount of live objects, (2) the runtime overhead is less than 20%, and (3) the maximum pause time caused by the compactor is comparable to that caused by the snapshot collector.

show abstract

Section: Space Overheadmentioning

confidence: 95%

Section: Related Workmentioning

confidence: 99%

Section: Pause Timementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Replication-Based Incremental Compaction

Ugawa

Yasugi

Yuasa

2008

2008 11th IEEE International Symposium on Object and Component-Oriented Real-Time Distributed Computing (ISORC)

View full text Add to dashboard Cite

show abstract

“…However, full compaction is costly since a substantial fraction of the objects may be moved and all references need to be updated [1,9,10]. Therefore, modern systems tend to either use compaction infrequently, or employ partial compaction, moving some objects in an attempt to reduce fragmentation and keep its cost acceptable [2,3,7,12,13]. Of course, the larger the space that is being compacted, the less fragmented the heap becomes; on the other hand, the overhead that is posed on the executing program increases.…”

Section: Introductionmentioning

confidence: 99%

Space overhead bounds for dynamic memory management with partial compaction

Bendersky

Petrank

2012

ACM Trans. Program. Lang. Syst.

View full text Add to dashboard Cite

Dynamic memory allocation is ubiquitous in today's runtime environments. Allocation and de-allocation of objects during program execution may cause fragmentation and foil the program's ability to allocate objects. Robson has shown that a worst case scenario can create a space overhead within a factor of log n of the space that is actually required by the program, where n is the size of the largest possible object. Compaction can eliminate fragmentation, but is too costly to be run frequently. Many runtime systems employ partial compaction, in which only a small fraction of the allocated objects are moved. Partial compaction reduces some of the existing fragmentation at an acceptable cost. In this paper we study the effectiveness of partial compaction and provide the first rigorous lower and upper bounds on its effectiveness in reducing fragmentation at a low cost.

show abstract

An efficient parallel heap compaction algorithm

Abuaiadh

Ossia

Petrank

et al. 2004

Proceedings of the 19th Annual ACM SIGPLAN Conference on Object-Oriented Programming, Systems, Languages, and Applications

View full text Add to dashboard Cite

We propose a heap compaction algorithm appropriate for modern computing environments. Our algorithm is targeted at SMP platforms. It demonstrates high scalability when running in parallel but is also extremely efficient when running single-threaded on a uniprocessor. Instead of using the standard forwarding pointer mechanism for updating pointers to moved objects, the algorithm saves information for a pack of objects. It then does a small computation to process this information and determine each object's new location. In addition, using a smart parallel moving strategy, the algorithm achieves (almost) perfect compaction in the lower addresses of the heap, whereas previous algorithms achieved parallelism by compacting within several predetermined segments.Next, we investigate a method that trades compaction quality for a further reduction in time and space overhead. Finally, we propose a modern version of the two-finger compaction algorithm. This algorithm fails, thus, re-validating traditional wisdom asserting that retaining the order of live objects significantly improves the quality of the compaction.The parallel compaction algorithm was implemented on the IBM production Java Virtual Machine. We provide measurements demonstrating high efficiency and scalability. Subsequently, this algorithm has been incorporated into the IBM production JVM.

show abstract

An algorithm for parallel incremental compaction

Cited by 8 publications

References 9 publications

Replication-Based Incremental Compaction

Replication-Based Incremental Compaction

Space overhead bounds for dynamic memory management with partial compaction

An efficient parallel heap compaction algorithm

Contact Info

Product

Resources

About