Global garbage collection for distributed heap storage systems

Ali, Khayri A. M.; Haridi, Seif

doi:10.1007/bf01414462

Cited by 11 publications

(8 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(2,3) It is difficult to compare the effectiveness of the two types of collector without implementations of both on a range of machines. However, it is possible to make general comments on the suitability of each type for particular classes of architecture.…”

Section: Asynchronous and Synchronous Collectionmentioning

confidence: 99%

A multicomputer garbage collector for a single-assignment language

Foster

1989

Int J Parallel Prog

View full text Add to dashboard Cite

An asynchronous garbage collector for a message-passing multiprocessor (multicomputer) is described. This combines Weighted Reference Counting (WRC) interprocessor collection and tracing intraprocessor collection to permit individual processors to reclaim local storage independently. A novel feature is the integration of Weighted Reference Counting collection and the communication algorithms required to support a global address space in a single assignment language. This significantly reduces communication overhead and space requirements attributable to garbage collection. In addition, techniques are described that avoid the creation of cyclic structures that cannot be reclaimed using WRC. Experimental studies performed in a concurrent logic programming system that incorporates the collector confirm its efficiency and the benefits of integrating garbage collector and language implementation.

show abstract

Section: Asynchronous and Synchronous Collectionmentioning

confidence: 99%

A multicomputer garbage collector for a single-assignment language

Foster

1989

Int J Parallel Prog

View full text Add to dashboard Cite

show abstract

“…Similar schemes have been used in some distributed systems [Ali85,Hug85], although these systems rely on global marking to collect both acyclic and cyclic interpartition garbage. Other schemes either propagate global marks separately from regular partition traces [JJ92], or are not guaranteed to terminate correctly in the presence of concurrent mutations [LQP92].…”

Section: Collecting Cyclic Garbagementioning

confidence: 99%

Partitioned garbage collection of a large object store

Maheshwari¹,

Liskov²

1997

Proceedings of the 1997 ACM SIGMOD International Conference on Management of Data - SIGMOD '97

View full text Add to dashboard Cite

This paper describes a new garbage collection scheme for large persistent object stores that makes efficient use of the disk and main memory. The heap is divided into partitions that are collected independently using information about inter-partition references. We present efficient techniques to maintain this information stably using auxiliary data structures in memory and the log. The result is a scheme that truly preserves the localized and scalable nature of partitioned collection.Remembering inter-partition references does not collect garbage cycles that span partitions. We describe a new global marking scheme that collects such garbage. We believe that it is the first scheme that piggybacks global marking on partitioned collection, does not delay the collection of acyclic garbage, and is guaranteed to terminate correctly in the presence of concurrent mutations. Further, it preserves the disk-efficient nature of our collector.We have implemented the part of garbage collection responsible for maintaining information about inter-partition references. We present a benchmark to evaluate this work and give performance results to show the advantages of our scheme.

show abstract

“…The local collection is based on tracing from a root set that includes the local persistent root objects as well as local objects that are referenced from other nodes, called the secondary roots. Different variants of distributed reference counting employ different methods and information to track the secondary roots, ranging from one-bit counts [Ali84,JJ92] to weighted reference counts [Bev87] to reference lists, in which each node tracks the identities of the nodes that refer to its objects [Bis77,SDP92,BENOW93].…”

Section: The Problem Contextmentioning

confidence: 99%

“…Some systems periodically invoke global marking to collect cyclic garbage [Ali84,JJ92]. Lang et al proposed marking within groups of nodes such that each round can tolerate failures of nodes outside the group [LQP92].…”

Section: Related Workmentioning

confidence: 99%

“…This research was supported in part by the Advanced Research Projects Agency of the Department of Defense, monitored by the Office of Naval Research under contract N00014-91-J-4136. using a complementary marking scheme to collect cyclic garbage [Ali84,JJ92,LQP92], or by migrating objects so that cyclic garbage ends up in a single node and is collected by the local collector [Bis77,SGP90,GF93]. The advantage of migration is that, like distributed reference counting, it is decentralized and fault-tolerant.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Collecting cyclic distributed garbage by controlled migration

Maheshwari

Liskov

1997

Distributed Computing

View full text Add to dashboard Cite

Distributed reference counting provides timely and faulttolerant garbage collection in large distributed systems, but it fails to collect cyclic garbage distributed across nodes. A common proposal is to migrate all objects on a garbage cycle to a single node, where they can be collected by the local collector. However, existing schemes have practical problems due to unnecessary migration of objects. We present solutions to these problems: our scheme avoids migration of live objects, batches objects to avoid a cascade of migration messages, and short-cuts the migration path to avoid multiple migrations. We use simple estimates to detect objects that are highly likely to be cyclic garbage and to select a node to which such objects are migrated. The scheme has low overhead, and it preserves the decentralized and fault-tolerant nature of distributed reference counting and migration.

show abstract

Global garbage collection for distributed heap storage systems

Cited by 11 publications

References 9 publications

A multicomputer garbage collector for a single-assignment language

A multicomputer garbage collector for a single-assignment language

Partitioned garbage collection of a large object store

Collecting cyclic distributed garbage by controlled migration

Contact Info

Product

Resources

About