This project is a CPU Scheduling Simulator that implements various scheduling algorithms, including:
- First-Come, First-Served (FIFO)
- Shortest Job First (SJF)
- Shortest Time to Completion First (STCF)
- Round Robin (RR)
The simulator processes jobs with different arrival times, execution times, and potential interruptions. It calculates metrics such as turnaround time and response time to compare scheduling strategies.
- Job Handling: Models job execution with arrival times, execution times, and I/O interruptions.
- Multiple Scheduling Algorithms: FIFO, SJF, STCF, and Round Robin scheduling.
- Performance Analysis: Calculates turnaround time, response time, and maximum completion time.
- Configurable Quantum Time: For Round Robin scheduling.
- Python 3.x
- NumPy
git clone <repo-url>
cd <project-folder>pip install numpyRun the main script:
python main.pyThe program will execute scheduling algorithms and output performance metrics.
│── main.py # Entry point for the simulation
│── operating_sys.py # Defines Operating System and Job classes
│── scheluder.py # Implements scheduling algorithms
└── README.md # Project documentationThis project is released under the MIT License.