Abstract

Hybrid Model for Concurrency Control of Transactions in Distributed Database Management System in the banking sector, where database integrity and consistency are very important was presented. Considering the four concurrency problems; Unrepeatable Read, Inconsistent Analysis, Lost Update and Phantom Read, Concurrency Control Methods has remained a crucial aspect in Distributed Database Management System (DDBMS). This research investigated two types of concurrency control methods; Pessimistic and Optimistic methods. While Pessimistic Concurrency control method uses locks to prevent deadlocks, Optimistic Concurrency Control method reduces lock contention and increase the risk of rollbacks. A hybrid model that integrated the two types of concurrency control methods to co-exist for faster and efficient concurrency control of transactions was developed. The hybrid model utilised the Genetic Algorithm (GA) for deep neural search on congested network in the distributed database and was equally used to optimise concurrent schedule by using its fitness function to measure deadlocks, performance and degree of serializability during transactions in an operation sequence. The Object-Oriented Analysis and Design (OOAD) methodology was adopted for the system design and the system was implemented using C# programming language. The developed model was evaluated using Confusion Metrics to check for the percentage of correctness, speed and performance. The hybrid model indicated that it outperformed individual Pessimistic or Optimistic methods by reducing transaction delays, avoiding deadlocks and improving overall system performance. The hybrid model enhanced the efficiency and reliability of banking transactions in distributed database system.

Keywords

  • Pessimistic
  • Optimistic
  • Concurrency
  • Distributed Database
  • Hybrid
  • Transactions

References

  1. Abduljalil E., Thabit F., Can O., Patil P. R. & Thorat, S. B. (2022). A New Secure 2PL Real-time Concurrency Control Algorithm (ES2PL). International Journal of Intelligent Networks, 3, 48-57.
  2. Asadi, M. & Khanbabaie, M. (2010). Developed S2PL in Transaction Concurrency with Emphasis on Deletion of Convoy Phenomenon and Improvement of Starvation of Transactions. Australian Journal of Basic and Applied Sciences (AJBAS), 4(10): 4682-4690.
  3. Bara F., Saadi C., & Chaoui H. (2020). Concurrency Control in Distributed Database. Laboratory for Systems Analysis, Information Processing and Industrial Management, Systems Engineering Laboratory. Accessed from https://39www.easychair.org/publications/preprint/82kw.
  4. Gabriel, B. C., Ojekudo, N. & Gabriel, M. N. (2021). Concurrency Control Technique for Transaction Processing in Distributed Database System Using Hybridized Model. International Journal of Advances in Engineering and Management), 3(7), 4122-4132.
  5. Gupta, M. K., Arora, R. K. & Bhati, B. Singh. (2018). Study of Concurrency Control Techniques in Distributed DBMS. International Journal of Machine Learning and Networked Collaborative Engineering, 2(4), 180–187.
  6. Quasim, T. (2013). An Efficient Approach for Concurrency Control in Distributed Database System. Indian Streams Research Journal (ISRJ), 3(9), 1-8.
  7. Shams, M., & Abohany, A. (2022). A Review on Concurrency Control Techniques in Database Management Systems. Kafrelsheikh Journal of Information Sciences (KJIS), 3(1), 1-10.
  8. Shebka, N. (2022). A Two-part Multi-Algorithm Concurrency Control Optimization Strategy for Distributed Database Systems. International Journal of Advanced and Applied Sciences (IJAAS), 9(7), 159-171.
  9. Yu, Z., Zuo, Y. & Xiong W. C. (2019). Concurrency Bug Avoiding Based on Optimized Software Transactional Memory. Scientific Programming, (4):1-19.