Pushing C# to the limit
summary
Builds a production-ready high-speed communications library in C# that uses shared memory lock-free pipes and unsafe pointers via memory-mapped views, plus interprocess signaling and a safe-disposable wrapper to avoid dangling pointers. Implements remoting-style expression-tree proxying with client-activated proxies, binary-writer fast paths and a BinaryFormatter fallback, and fully asynchronous Task-based calls. Achieves about ~560k msgs/sec in a stress test and reports a 5–10x latency improvement over .NET remoting, and the system works reliably across processes.
AI-written summary · about
chapters (4)
related talks
27mC# as a system language
01h 06m.NET Systems Programming Learned the Hard Way
01h 04mTurbocharged: Writing High-Performance C# and .NET Code
01h 14mTurbocharged: Writing High-performance C# and .NET code
40mReal-world examples on optimizing .NET performance
01h 01mTurbocharged: Writing High-Performance C# and .NET Code
01h 01mBoosting memory management in interoperability scenarios
01h 09mBuild a Highly Performant Interservice Communication with gRPC for ASP NET Core
