MSDN Documentation

Understanding Garbage Collection in .NET

Garbage collection (GC) is an automatic memory management feature in the .NET Common Language Runtime (CLR). It reclaims memory occupied by objects that are no longer in use by the application, preventing memory leaks and simplifying development.

How Garbage Collection Works

The GC operates in cycles. When the CLR detects that the managed heap is running low on memory, it initiates a garbage collection process. This process involves the following steps:

1. Marking: The GC starts from the "root set" (which includes static objects, thread stacks, and objects referenced by the CPU registers) and traverses the object graph. It marks all objects that are reachable from the root set.
2. Sweeping: After marking, the GC iterates through the managed heap. Any object that was not marked is considered unreachable and its memory is reclaimed.
3. Compacting (Optional): To combat fragmentation, the GC can move the surviving (marked) objects closer together. This creates larger contiguous blocks of free memory, improving allocation performance.

Generations

To optimize the collection process, .NET uses a generational garbage collector. This approach is based on the observation that most objects have a short lifespan. The managed heap is divided into three generations:

• Generation 0: Contains newly allocated objects. This generation is collected most frequently.
• Generation 1: Contains objects that survived one or more collections of Generation 0.
• Generation 2: Contains long-lived objects that have survived multiple collections. This generation is collected the least frequently.

When a collection occurs, the GC focuses on the lowest-numbered generation that contains objects. If Generation 0 is collected and a significant amount of memory is reclaimed, the GC might stop. Otherwise, it will proceed to collect Generation 1, and so on.

Types of Garbage Collectors

.NET offers different GC modes to suit various application needs:

• Workstation GC: Optimized for client applications, providing lower latency. It runs in a single thread.
• Server GC: Optimized for server applications, maximizing throughput by using multiple GC threads running in parallel.

Configuration and Best Practices

While the GC is largely automatic, understanding its behavior can help you write more efficient code.

• Minimize Allocations: Frequent object creation can put a strain on the GC. Consider object pooling for frequently used objects.
• Dispose of Unmanaged Resources: The GC only manages managed memory. For unmanaged resources (like file handles, database connections), implement the IDisposable interface and use using statements to ensure they are released promptly.
• Avoid Long-Lived References: Keeping references to objects longer than necessary can prevent them from being collected.

GC in .NET Core and later

While the core concepts remain the same, .NET Core and subsequent versions have seen continuous improvements to the GC, including enhanced performance and scalability.