<![CDATA[systems that directly affects system performance and stability, especially in multitasking environments. Differences in memory management mechanisms can result in varying performance characteristics in terms of memory allocation, fragmentation, and system management overhead. This study aims to analyze and compare the performance of paging, segmentation, and the combination of paging-segmentation mechanisms thru a simulation approach. The method used is event-driven simulation on a multitasking operating system with a single processing unit (single CPU). Performance evaluation is conducted based on the success rate of process allocation, memory fragmentation, and page fault frequency as indicators of memory management overhead. Simulation results show that the paging mechanism can maintain a high success rate of process allocation across all system load scenarios, but with an increase in page fault frequency. The segmentation mechanism shows optimal performance under low load, but experiences a decrease in allocation success under high load due to external fragmentation. Meanwhile, the paging-segmentation combination mechanism provides the most stable performance with a consistent allocation success rate and more controlled memory fragmentation. The results of this study indicate that the combination mechanism is the most balanced approach for multitasking systems with dynamic workload variations.]]>
Copyrights © 2026
