Breaking Deadlocks: A Novel Method for Dynamic Deadlock Resolution in Concurrent Systems
DOI:
https://doi.org/10.37591/joasp.v10i3.765Keywords:
Deadlock handling, concurrent systems, dynamic deadlock resolution, proactive and reactive strategies, system adaptabilityAbstract
It is known that deadlock is one of the major challenges in concurrent systems, as sometimes this deadlock may lead to system failure as well as performance degradation. Traditional deadlock handling methods, for example, deadlock detection and prevention, have several limitations for systems that have a dynamic nature in which the structure of the system and the resource requirements change over time. In this article, we present a novel technique that can eliminate deadlocks and overcome the aforementioned drawbacks by using both proactive and reactive techniques. The approach we proposed is unique in that it introduces a dynamic deadlock resolution mechanism that dynamically adapts to changing system conditions, thus breaking the possibly growing deadlocks while not compromising system responsiveness and resource utilization. The findings highlight the effectiveness and scalability of their use in solving deadlock issues related to different concurrent systems.
References
Eswaran KP, Gray JN, Lorie RA, Traiger IL. The notions of consistency and predicate locks in a database system. Commun ACM. 1976; 19 (11): 624–633.
Birman KP, Joseph TA. Exploiting virtual synchrony in distributed systems. ACM Trans Comput Syst. 1987; 5 (4): 320–347.
Abadi M, Lamport L. The existence of refinement mappings. Theor Comput Sci. 1993;82(2):253–284.
Agarwal R, Meyer D. Deadlock-free message routing in processor networks with wormhole routing. IEEE Trans Parallel Distrib Syst. 1996; 7 (3): 261–274.
Srivastava A, Chatterjee S. Dynamic deadlock prevention in distributed real-time systems. IEEE Trans Parallel Distrib Syst. 2001; 12 (6): 601–619.
Elnozahy ENM, Alvisi L, Wang YM, Johnson DB. A survey of rollback-recovery protocols in message-passing systems. ACM Comput Surv. 2002; 34 (3): 375–408.
Yu H, Prasanna VK. Deadlock-free and livelihood-ensuring routing in irregular networks. IEEE Trans Parallel Distrib Syst. 2003; 14 (9):896–910.
Kung HT, Robinson JT. Deadlock-free adaptive mesh routing in chip multiprocessors. IEEE Trans Comput. 2009;58(1):48–62.
Zhang F, Butt AR. A survey of deadlock prevention techniques in interconnection networks. IEEE Trans Parallel Distrib Syst. 2014; 25 (4): 981–994.
Blinkov Y, Garzaran MJ, Padua D, Pouchet L. Scalable deadlock detection for large-scale parallel applications. ACM Trans Parallel Comput. 2018; 5 (1): Article 1.
Rakamaric Z, Vechev MT. A decision procedure for validating livelock freedom of concurrent programs. ACM SIGPLAN Not. 2018;53(1):184–198.
Bouzida A, Laidi B. A new approach for deadlock detection in distributed systems based on dynamic analysis. J Parallel Distrib Comput. 2020; 145: 43–52.
Liu Y, Wang X, Zhang Y, Zhang P. A dynamic resource allocation algorithm for deadlock prevention in edge computing. IEEE Trans Ind Informatics. 2020;16(10):6586–6594.
Chen Y, Feng X, Zhang Y. A novel deadlock detection and recovery approach for concurrent systems. Inf Sci. 2021; 572: 76–90.
Rizvi SAH, Islam N. A machine learning-based deadlock detection mechanism for concurrent systems. Neural Comput Appl. 2021; 33 (16): 8049–8063.
Xu Y, Guo W, Zhao J. A probabilistic deadlock detection method based on reinforcement learning in edge computing. IEEE Trans Ind Informatics. 2021; 17 (3):2032–2041.
Suresh M, Yadav S. Dynamic deadlock avoidance algorithm using hybrid approach for resource allocation in multicore systems. J Parallel Distrib Comput. 2022; 163: 29–42.
Zhang H, Lu Y, Li C, Zhou X. A novel adaptive deadlock resolution mechanism for real-time embedded systems. ACM Trans Embedded Comput Syst. 2022; 21 (1): Article 10.
Chen C, Gupta RK. Distributed deadlock detection and resolution in IoT edge computing systems. IEEE Trans Parallel Distrib Syst. 2023; 34(2): 507–520.
Wang J, Jin L, Yang L. Adaptive deadlock prevention algorithm for task scheduling in fog computing. Future Gener Comput Syst. 2023; 131: 151–160.
