Overview
Memory management is the process of controlling and coordinating computer memory, assigning portions (blocks) to various running programs to optimize overall system performance.
Core Concepts
- The Stack:
- Stores local variables and function call frames.
- Fast access, automatic allocation/deallocation.
- Limited size (can lead to Stack Overflow).
- The Heap:
- Stores objects, arrays, and dynamically allocated memory.
- Larger size, slower access.
- Requires manual or automatic management.
- Garbage Collection (GC): An automatic memory management process that finds and deletes objects that are no longer reachable by the program.
- Mark-and-Sweep: The most common GC algorithm.
- Reference Counting: Tracks how many references point to an object.
- Memory Leaks: Occur when memory that is no longer needed is not released back to the system.
Code Examples
// Potential Memory Leak in JS
let leak = [];
setInterval(() => {
leak.push(new Array(1000).fill("leak")); // Growing heap memory
}, 100);
// Proper cleanup
function cleanup() {
leak = []; // Allow GC to reclaim the memory
}Use Cases
- Game Development: Manual memory management (C++) is used to avoid “GC spikes” (stutters) during gameplay.
- Embedded Systems: Strict memory limits require precise control over every byte.
- High-Performance Computing: Optimizing “cache locality” to reduce memory access times.
Gotchas
- Circular References: In simple reference-counting GC, two objects pointing to each other will never be deleted, even if the rest of the app can’t reach them.
- Assuming GC is “Free”: Garbage collection consumes CPU cycles. In high-throughput systems, tuning the GC is a critical task.