Zero-Copy C# with Span

Modern .NET is fast because of Span. It allows you to look at a piece of data (in a string, an array, or even unmanaged memory) without Copying it. This is the secret to high-throughput parsers and serializers.

1. The Problem with Substring()

When you call `str.Substring(0, 10)`, C# creates a WHOLE NEW string in memory. If you are parsing a 1GB file, you will create 1GB of temporary garbage strings. **Span** points to the original memory—it's like a 'Window' into the data.

2. Ref Struct Constraints

Span is a Ref Struct, which means it can ONLY live on the stack. You cannot use it as a field in a class or use it in an `async` method. For those cases, you use **Memory**, which is the heap-safe version of Span.

4. Interview Mastery

Q: "How did Span make .NET Core faster than .NET Framework?"

Architect Answer: "Span allowed the .NET team to rewrite core libraries (like JSON, HTTP, and Kestrel) to use Zero-Allocation patterns. Instead of copying bytes from the network card to an array, and then to a string, and then to a DTO, Span allows us to parse the data directly from the network buffer. This reduced allocations by 90% and is why .NET is now one of the fastest web frameworks in the world."