Mini Runtime Allocator

This is a simple memory allocator simulator written in C++. It mimics a heap allocator with basic allocation and deallocation functions, simulating a contiguous memory region. The project is designed to demonstrate how memory allocation strategies work in a runtime system.

Features

• Fixed-size memory (default: 1024 bytes)

• Allocation strategies:

  • First-Fit
  • Best-Fit
  • Worst-Fit
  • Buddy System (power-of-two splitting & merging)

• Command-line interface (CLI)

• Supports alloc, free, show, strategy, stats, visual, benchmark, and exit commands

• Benchmarking framework with CSV output for analysis

• ASCII visualization of memory layout

Usage

Build (with CMake)

mkdir build
cd build
cmake ..
make

Run

./src/runtime

CLI Commands

Command	Description
alloc <size>	Allocate memory block of given size
free <id>	Free block by allocation ID
show	Show current memory layout
strategy <name>	Switch strategy to first, best, worst, or buddy
stats	Show fragmentation statistics
visual	Show ASCII visualization of memory (used = #, free = .)
benchmark	Run benchmark for all strategies, output CSV + summary
exit	Exit the program

Example

> alloc 300
Allocated ID: 1
> alloc 200
Allocated ID: 2
> show
[0 - 299] Used (ID: 1) Size: 300  
[300 - 499] Used (ID: 2) Size: 200  
[500 - 1023] Free Size: 524  
> visual
[ASCII Memory Map]
###.................................................
> benchmark
[Benchmark Finished] Strategy=first Ops=1000 Time=12 ms
[Benchmark Finished] Strategy=best Ops=1000 Time=13 ms
[Benchmark Finished] Strategy=worst Ops=1000 Time=12 ms
[Benchmark Finished] Strategy=buddy Ops=1000 Time=9 ms
Results saved to benchmark_first.csv, benchmark_best.csv, benchmark_worst.csv, benchmark_buddy.csv

Benchmark & Visualization

• Benchmark generates CSV files for all strategies:

  • benchmark_first.csv
  • benchmark_best.csv
  • benchmark_worst.csv
  • benchmark_buddy.csv

• Each CSV includes:

  • step, total_free, max_free, fragments, fragmentation_ratio

• A Python script plot_benchmark.py is provided in the project root to compare fragmentation ratios:

python3 plot_benchmark.py

This generates benchmark_comparison.png, plotting fragmentation ratio vs operations for all strategies. It also prints average fragmentation ratio for each strategy in the console.

Testing

This project uses Catch2, a modern C++ testing framework. After building, you can run tests with:

ctest --verbose

Future Extensions

• Extended fragmentation analysis (internal + external)
• Advanced allocators (e.g., Slab Allocator, TLSF)
• Interactive/graphical visualizer

---

MIT License © 2025