Scheduling Algorithm in Minix 3.2.1

Academic Project

Overview

Implemented advanced process scheduling algorithms directly in the Minix 3.2.1 operating system kernel, exploring real-time and probabilistic scheduling paradigms.

Key Contributions

• Earliest Deadline First (EDF): Implemented optimal real-time scheduling algorithm for deadline-sensitive tasks
• Lottery Scheduling: Developed probabilistic scheduling mechanism providing proportional share resource allocation
• Kernel Integration: Modified kernel space code to integrate custom scheduling policies
• Performance Analysis: Evaluated scheduling algorithm performance under different workload conditions

Technical Skills Demonstrated

• Operating Systems: Deep kernel-level programming and system call implementation
• Real-time Systems: Understanding of deadline constraints and timing guarantees
• Algorithm Implementation: Translation of theoretical scheduling algorithms to practical implementations
• System Programming: Low-level C programming in kernel space with strict resource constraints

Learning Outcomes

This project provided comprehensive understanding of operating system internals, process management, and the critical role of scheduling in system performance and real-time guarantees.