تحليل محسن لأداء خوارزميات التوجيه في شبكات الحاسوب باستخدام محاكاة أحمال حركة مرور تكيفية

This paper compares Dijkstra’s and Bellman–Ford algorithms in computer networks using a Monte Carlo simulation with variable traffic loads. A fully connected directed network was implemented in Python with randomly assigned positive link weights. Low, medium, and high traffic conditions were evaluated using delay, throughput, packet loss, and path cost. Repeated simulation runs were used to examine variability and estimate uncertainty. The reported results show increasing delay and packet loss and decreasing throughput as traffic intensity rises. Both algorithms display broadly similar performance across the three load levels, with small differences in delay and throughput. These similarities are consistent with both algorithms solving the same shortest-path problem on the modeled network. The framework supports controlled comparisons under changing load assumptions and illustrates how traffic intensity can dominate modeled performance. However, the delay, throughput, and loss measures are generated by simplified formulas rather than a packet-level network model. The findings therefore describe the specified simulation and should not be treated as evidence of superior congestion control or routing convergence in operational networks. Further evaluation using realistic topologies, explicit queueing, and paired simulation runs would strengthen the comparison.