Analisis Banker’s Algorithm untuk Penghindaran Deadlock Berbasis Simulasi Kuantitatif Multiskenario

Christian Bastanta Sembiring Meliala (Universitas Pamulang, Tangerang Selatan)
Teti Desyani (Universitas Pamulang, Tangerang Selatan)
Moch Ibba Ali Yassin (Universitas Pamulang, Tangerang Selatan)
Aldiansyah Sastrawinata (Universitas Pamulang, Tangerang Selatan)
Mikael Surya Saputra (Universitas Pamulang, Tangerang Selatan)
Brian Aidil Rizkita (Universitas Pamulang, Tangerang Selatan)
Rizki Arohman Maulana (Universitas Pamulang, Tangerang Selatan)

TIN: TERAPAN INFORMATIKA NUSANTARA
Vol 7 No 1 (2026): June 2026

Deadlock represents a critical threat in operating system resource management, as it has the potential to bring all computational processes to a complete halt. This study examines the effectiveness, efficiency, and constraints of the Banker's Algorithm as a deadlock avoidance mechanism through a multi-scenario quantitative simulation. The data were derived from simulations involving three core components: the resource allocation matrix (Allocation), the maximum process requirement declaration (Max), and the resource availability vector (Available), within a system configuration consisting of five processes and three resource types. The findings demonstrate that the Banker’s Algorithm accurately distinguishes between safe and unsafe states through its two primary mechanisms: the Safety Algorithm and the Resource-Request Algorithm. With Available set to [3, 3, 2], the algorithm successfully identified the safe execution sequence ⟨P1, P3, P4, P0, P2⟩, ensuring all processes could complete without deadlock risk. When Available was reduced to [2, 1, 0], the system entered an unsafe state in which no process could initiate execution. Through multi-scenario simulations, the critical transition threshold from a safe to an unsafe state was identified at approximately 83% resource utilization. In terms of efficiency, the O(n²×m) time complexity makes the algorithm well-suited for small to medium-scale systems, though it may become a performance bottleneck in large-scale cloud computing environments. This study produces a quantitative evaluation framework that can serve as a reference for implementing the Banker’s Algorithm in modern operating systems.