Performance tricks I learned from contributing to open source .NET packages
summary
Contributions to the Azure .NET SDK drive concrete wins: 80+ PRs applied various micro‑optimisations that cut GC and raised throughput — replacing LINQ/Any with loops yields 20–40% fewer GCs and 20–40% more throughput, a collection‑based refactor adds up to 5–64% extra throughput and 23–61% GC reduction, removing closure allocations and switching to ValueTask patterns improves a pipeline by 74–78%, and a span/stackalloc + ArrayPool partition‑hash rewrite gains 38–47%; use BenchmarkDotNet and allocation tools (Rider/VS) to measure.
AI-written summary · about
chapters (8)
- 0:00Performance optimizations in .NET
- 2:13Focus on C# and .NET
- 7:32Reducing GC overhead
- 21:14Avoiding lambda allocations
- 31:06Buffer pooling efficiently
- 39:52Span for memory copying
- 53:20Applying optimizations
- 57:38Q&A
related talks
01h 23mPerformance Tricks I Learned From Contributing to Open-Source .NET Packages
47mUnlocking Performance Improvements in .NET
01h 04mTurbocharged: Writing High-Performance C# and .NET Code
46mDeep dive into .NET performance and native AOT
01h 14mTurbocharged: Writing High-performance C# and .NET code
01h 01mTurbocharged: Writing High-Performance C# and .NET Code
56mState of the .NET Performance
59mWriting (in)efficient C#