A biological-like memory allocation scheme using simulation
Gasydech Lergchinnaboot, Peraphon Sophatsathit · 2017
This research proposes a novel memory allocation scheme to efficiently handle memory management. The scheme employs biological behavioral principles of the life forms. At the principal construct lives the cells having limited resources, yet passively operates with little overhead. The proposed scheme imitates this unicellular characterization to execute one task at a time using First-In-First-Out queue. Execution is regulated by a global clock that permits one active task at any given time in memory. Consequently, low overhead memory allocation can be achieved without the need for elaborate scheduling and other supporting algorithms. The most anticipatory benefit is simplicity that permits straightforward technological transfer of the proposed scheme to hardware. The contributions are to systematically mitigate the memory wall and reduce power consumed by memory management activities.