Performance of memory reclamation for lockless synchronization

Hart, Tae L.; McKenney, Paul E.; Brown, Angela Demke; Walpole, Jonathan

doi:10.1016/j.jpdc.2007.04.010

Cited by 134 publications

(122 citation statements)

References 29 publications

Supporting

Mentioning

121

Contrasting

Order By: Relevance

“…OPTIK decouples concurrency control from memory reclamation. Accordingly, OPTIK can be used with practically any memory-reclamation scheme, such as hazard pointers [38], RCU [35], quiescent states [19,20]. Our CSDS implementations with OPTIK use ssmem, 8 a simple memory allocator with quiescent-based memory reclamation.…”

Section: Practical Considerationsmentioning

confidence: 99%

“…The delete operation (Figure 8(a)) is the most complex operation of the OPTIK-based linked list, because it requires locking two nodes; the one being deleted and its predecessor node. Traversing the list (lines 11-15) keeps track of these two version numbers that are later used for locking with optik trylock version (lines [18][19][20][21][22][23]. If locking the predecessor node fails, the operation is restarted, otherwise the node to be deleted is locked.…”

Section: Optik-based Linked Listmentioning

confidence: 99%

See 1 more Smart Citation

Optimistic concurrency with OPTIK

GuerraouiRachid

TrigonakisVasileios

2016

SIGPLAN Not.

View full text Add to dashboard Cite

We introduce OPTIK, a new practical design pattern for designing and implementing fast and scalable concurrent data structures. OPTIK relies on the commonly-used technique of version numbers for detecting conflicting concurrent operations. We show how to implement the OPTIK pattern using the novel concept of OPTIK locks. These locks enable the use of version numbers for implementing very efficient optimistic concurrent data structures. Existing state-of-the-art lock-based data structures acquire the lock and then check for conflicts. In contrast, with OPTIK locks, we merge the lock acquisition with the detection of conflicting concurrency in a single atomic step, similarly to lock-free algorithms. We illustrate the power of our OPTIK pattern and its implementation by introducing four new algorithms and by optimizing four state-of-the-art algorithms for linked lists, skip lists, hash tables, and queues. Our results show that concurrent data structures built using OPTIK are more scalable than the state of the art.

show abstract

Section: Practical Considerationsmentioning

confidence: 99%

Section: Optik-based Linked Listmentioning

confidence: 99%

Optimistic concurrency with OPTIK

GuerraouiRachid

TrigonakisVasileios

2016

SIGPLAN Not.

View full text Add to dashboard Cite

show abstract

“…Quiescence-based techniques [20,14,21] such as RCU [36] have threads accessing the data structure register the start and end of methods, and have the reclaiming thread wait a sufficiently long period of time until it can be ensured that no registered thread holds a reference to a deleted object. These techniques are efficient since here is no per-read tracking, making them compatible with invisible traversals.…”

Section: Concurrent Memory Reclamationmentioning

confidence: 99%

ThreadScan

Alistarh

Leiserson

Matveev

et al. 2015

Proceedings of the 27th ACM Symposium on Parallelism in Algorithms and Architectures

View full text Add to dashboard Cite

The concurrent memory reclamation problem is that of devising a way for a deallocating thread to verify that no other concurrent threads hold references to a memory block being deallocated. To date, in the absence of automatic garbage collection, there is no satisfactory solution to this problem; existing tracking methods like hazard pointers, reference counters, or epoch-based techniques like RCU, are either prohibitively expensive or require significant programming expertise, to the extent that implementing them efficiently can be worthy of a publication. None of the existing techniques are automatic or even semi-automated.In this paper, we take a new approach to concurrent memory reclamation: instead of manually tracking access to memory locations as done in techniques like hazard pointers, or restricting shared accesses to specific epoch boundaries as in RCU, our algorithm, called ThreadScan, leverages operating system signaling to automatically detect which memory locations are being accessed by concurrent threads.Initial empirical evidence shows that ThreadScan scales surprisingly well and requires negligible programming effort beyond the standard use of Malloc and Free.

show abstract

“…One reason RCU is heavily used is that it eases lock-based programming when the locks themselves are dynamically created and destroyed, which occurs frequently in concurrent programs. However, RCU is not heavily used in applications, in part because prior user-level RCU-like algorithms severely constrained application design [7], incurred heavy read-side overhead [8,9], or relied on sequential consistency and garbage collection [10,11]. The popularity of RCU in operating-system kernels owes much to the fact that kernels can accommodate the global constraints imposed by the high-performance quiescent-state based reclamation (QSBR) class of RCU implementations.…”

Section: Introductionmentioning

confidence: 99%

“…The popularity of RCU in operating-system kernels owes much to the fact that kernels can accommodate the global constraints imposed by the high-performance quiescent-state based reclamation (QSBR) class of RCU implementations. QSBR implementations provide unmatched performance and scalability for read-mostly data structures on cache-coherent shared-memory multiprocessors [7], even with weakly ordered hardware and compilers. Whereas we cannot yet put forward a single user-level RCU implementation that is ideal for all environments, the three classes of RCU implementations described in this paper should suffice for most user-level uses of RCU.…”

Section: Introductionmentioning

confidence: 99%

User-Level Implementations of Read-Copy Update

Desnoyers¹,

McKenney²,

Stern³

et al. 2012

IEEE Trans. Parallel Distrib. Syst.

Self Cite

129

View full text Add to dashboard Cite

Abstract-Read-copy update (RCU) is a synchronization technique that often replaces reader-writer locking because RCU's read-side primitives are both wait-free and an order of magnitude faster than uncontended locking. Although RCU updates are relatively heavy weight, the importance of read-side performance is increasing as computing systems become more responsive to changes in their environments.RCU is heavily used in several kernel-level environments. Unfortunately, kernel-level implementations use facilities that are often unavailable to user applications. The few prior userlevel RCU implementations either provided inefficient readside primitives or restricted the application architecture. This paper fills this gap by describing efficient and flexible RCU implementations based on primitives commonly available to userlevel applications.Finally, this paper compares these RCU implementations with each other and with standard locking, which enables choosing the best mechanism for a given workload. This work opens the door to widespread user-application use of RCU.

show abstract

Performance of memory reclamation for lockless synchronization

Cited by 134 publications

References 29 publications

Optimistic concurrency with OPTIK

Optimistic concurrency with OPTIK

ThreadScan

User-Level Implementations of Read-Copy Update

Contact Info

Product

Resources

About