Advertisement

Optimal Recoverable Mutual Exclusion Using only FASAS

  • Prasad Jayanti
  • Siddhartha Jayanti
  • Anup Joshi
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11028)

Abstract

Recent research has focused on designing concurrent algorithms that are resilient to process crashes. The idea is to leverage non-volatile memory so that processes can recover from crashes with as little disruption to the normal behavior of the system as possible. We present the first Recoverable Mutual Exclusion algorithm whose Remote Memory Reference (RMR) complexity is optimal for both Cache-Coherent (CC) and Distributed Shared Memory (DSM) machines. If a process fails f times during its attempt to acquire the Critical Section, our algorithm ensures that the process incurs O(1) RMRs on a DSM machine and O(f) RMRs on a CC machine, which we prove is an optimal bound. Our algorithm improves on a recent algorithm by Golab and Hendler in three ways: It has a provably optimal RMR complexity, has a wait-free Exit section, and less reliance on instructions that are not commonly supported on multiprocessors. In particular, Golab and Hendler’s algorithm relies on hardware support for both Fetch-And-Store-And-Store (FASAS) and Double-Word Compare-And-Swap (DCAS), while our algorithm relies only on FASAS. (If X and Y are shared variables and v is a value, FASAS(XYv) writes X’s value in Y and writes v in X, all in a single atomic action.)

Keywords

Concurrent algorithm Synchronization Mutual exclusion Recovery Fault tolerance Non-volatile main memory Shared memory Multi-core algorithms 

Notes

Acknowledgment

We are grateful to the anonymous reviewers for their careful and detailed reviews.

References

  1. 1.
    Dijkstra, E.W.: Solution of a problem in concurrent programming control. Commun. ACM 8(9), 569 (1965)CrossRefGoogle Scholar
  2. 2.
    Golab, W., Hendler, D.: Recoverable mutual exclusion in sub-logarithmic time. In: Proceedings of the ACM Symposium on Principles of Distributed Computing. PODC 2017, pp. 211–220. ACM, New York (2017)Google Scholar
  3. 3.
    Golab, W., Ramaraju, A.: Recoverable mutual exclusion: [extended abstract]. In: Proceedings of the 2016 ACM Symposium on Principles of Distributed Computing. PODC 2016, pp. 65–74. ACM, New York (2016)Google Scholar
  4. 4.
    Jayanti, P., Joshi, A.: Recoverable FCFS mutual exclusion with wait-free recovery. In: 31st International Symposium on Distributed Computing. DISC, pp. 30:1–30:15 (2017)Google Scholar
  5. 5.
    Ramaraju, A.: RGLock: Recoverable mutual exclusion for non-volatile main memory systems. Master’s thesis. University of Waterloo (2015)Google Scholar
  6. 6.
    Craig, T.S.: Building FIFO and Priority-Queuing Spin Locks from Atomic Swap. Technical report TR-93-02-02, Department of Computer Science, University of Washington, February 1993Google Scholar
  7. 7.
    Dvir, R., Taubenfeld, G.: Mutual exclusion algorithms with constant RMR complexity and wait-free exit code. In: Proceedings of The 21st International Conference on Principles of Distributed Systems, OPODIS 2017 (2017)Google Scholar
  8. 8.
    Lamport, L.: A new solution of Dijkstra’s concurrent programming problem. Commun. ACM 17(8), 453–455 (1974)MathSciNetCrossRefGoogle Scholar
  9. 9.
    Mellor-Crummey, J.M., Scott, M.L.: Algorithms for scalable synchronization on shared-memory multiprocessors. ACM Trans. Comput. Syst. 9(1), 21–65 (1991)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Prasad Jayanti
    • 1
  • Siddhartha Jayanti
    • 2
  • Anup Joshi
    • 1
  1. 1.Dartmouth CollegeHanoverUSA
  2. 2.Massachusetts Institute of TechnologyCambridgeUSA

Personalised recommendations