Concurrent deferred reference counting with constant-time overhead

Anderson, D. Mark; Blelloch, Guy E.; Wei, Yuanhao

doi:10.1145/3453483.3454060

Cited by 13 publications

(3 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In Section 6, we apply the RC scheme of [2] to manage version list nodes as it adds only constant time overhead in expectation and it is the only one that maintains our desired time bounds. Epoch-based reclamation (EBR) [22,14] employs a counter that is incremented periodically and is used to divide the execution into epochs.…”

Section: Related Workmentioning

confidence: 99%

“…We now describe how to safely reclaim the nodes spliced out of version lists and the Descriptor objects that are no longer needed. We apply an implementation of Reference Counting (RC) [2] with amortized expected constant time overhead to a slightly modified version of our list. To apply this scheme to Figure 4, we add a reference count field to each Node or Descriptor and replace raw pointers to Nodes or Descriptors with reference-counted pointers.…”

Section: Memory Reclamation For Version Listsmentioning

confidence: 99%

“…We use a reference counting reclamation scheme and this requires modifying our doubly-linked list algorithm slightly to maintain the desired space bounds. We apply an existing concurrent reference counting implementation [2] that employs a local hash table per process which causes the time bounds of our reclamation to become amortized O(1) in expectation. It also requires an additional fetch-and-add instruction, whereas the rest of our algorithms require only read and CAS.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Space and Time Bounded Multiversion Garbage Collection

Ben-David,

Blelloch,

Fatourou

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

We present a general technique for garbage collecting old versions for multiversion concurrency control that simultaneously achieves good time and space complexity. Our technique takes only O(1) time on average to reclaim each version and maintains only a constant factor more versions than needed (plus an additive term). It is designed for multiversion schemes using version lists, which are the most common.Our approach uses two components that are of independent interest. First, we define a novel rangetracking data structure which stores a set of old versions and efficiently finds those that are no longer needed. We provide a wait-free implementation in which all operations take amortized constant time. Second, we represent version lists using a new lock-free doubly-linked list algorithm that supports efficient (amortized constant time) removals given a pointer to any node in the list. These two components naturally fit together to solve the multiversion garbage collection problem-the range-tracker identifies which versions to remove and our list algorithm can then be used to remove them from their version lists. We apply our garbage collection technique to generate end-to-end time and space bounds for the multiversioning system of Wei et al. (PPoPP 2021).

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Memory Reclamation For Version Listsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation