Bentest

Bentest is a Performance Monitoring Tool written in C. It measures runtime statistics of a given program. It repeatedly executes the specified program, collects metrics like execution time, CPU usage, and voluntary context switches, and outputs performance summaries to help identify potential inefficiencies. Think of it as Linux's time on steroids.

Currently supporting unix-based systems and planned to support windows soon.

Features

• Executes a program multiple times and measures:
  • User time
  • System time
  • Total execution time
  • CPU usage
  • Voluntary context switches
  • Involuntary context switches
  • Page Faults (soft and hard)
  • File System I/O (input and output) blocks
• Uses getrusage for high-accuracy CPU time measurement.
• Accepts command-line arguments for flexibility.
• Outputs statistics to help analyze program performance.

Requirements

• A Linux-based operating system. (Planning to make it portable)
• GCC (GNU Compiler Collection) to compile the source code.

Installation

1. Clone the repository:

   git clone <repository_url>
   cd <repository_name>

2. Build the tool using the provided Makefile:

   make

   This will create the bentest executable as build/bentest.

   Note: Ensure make is installed on your system.

Usage

Run the tool with the following command:

./build/bentest <num_iterations> <program> [args...]

Arguments:

• <num_iterations>: The number of times to execute the program. Must be a positive integer.
• <program>: The program to execute (e.g., gzip, curl, python scripts, etc.).
• [args...]: Additional arguments for the program.

Examples:

1. Compress a file with gzip:

   ./build/bentest 3 gzip sample.txt

   This runs gzip sample.txt three times and monitors its performance.

Output

The program outputs statistics for each run and a summary of:

• User time
• System time
• Total execution time
• CPU usage percentage
• Voluntary context switches (indicating potential system call frequency)

How It Works

1. Argument Parsing: The tool validates the number of iterations and the program to execute.
2. Process Creation: Uses fork() to create a child process for each execution.
3. Program Execution: The child process runs the program using execvp().
4. Timing and Metrics:
   • Captures execution time using clock_gettime
   • Collects resource usage data using getrusage
5. Statistics Calculation: Updates and aggregates runtime statistics across iterations.

Limitations

• Internal commands (e.g., cd, exit) are not supported directly.

Future Enhancements

• Add memory usage monitoring.
• Monitor system calls
• Improve the accuracy for total time measurement.
• Portability to non-unix systems.

Contributing

Contributions are welcome! If you find a bug or want to add a feature:

1. Fork the repository.
2. Create a new branch.
3. Submit a pull request with a detailed description.

License

This project is licensed under the MIT License.

---

Contact

For any inquiries or feedback, feel free to reach out via GitHub issues.

---

Happy monitoring! 🚀