High-performance C++20 distributed rate limiter using gRPC and Redis. Features a tiered L1/L2 architecture with lock-free SPSC queues and sharded hot-path caching, achieving 180k+ req/s with sub-300ns inter-thread latency.
A simple high-performance ring buffer handling 10,000,000 messages/sec.
ZFS tools, SMR drive tuning, and USB queue monitoring utilities
Proof of Concept demonstrating mechanical sympathy principles from Martin Fowler's article, applied to a low-latency order processing system.
A wait-free SPSC (Single-Producer Single-Consumer) lock-free queue. Achieving 225M+ ops/s through cache-line alignment and acquire-release semantics.
Zero-dependency browser experiments. WebAssembly edge. Service Worker mastery. No bundlers. No frameworks. Vanilla JS + Web APIs. Custom allocators, lock-free queues, SIMD via WASM, OffscreenCanvas. 60fps terminal emulators. SharedArrayBuffer. ReadableStream zero-copy. index.html → production.
A circuit-based programming framework and K2 compiler plugin. Writes like idiomatic Kotlin, synthesizes into wait-free, zero-allocation FSMs. Bridging the gap between JVM/Android abstractions and DPDK-tier hardware speeds.
A Java matching engine where one thread owns the book and orders reach it through a lock-free ring. Matching costs 0.73 bytes per command at 20M commands a second, and the raw JMH and HdrHistogram data is committed alongside.
Deep dives into Linux kernel internals for low-latency systems engineering — context switches, page faults, NUMA, and CPU isolation, explained from first principles with original diagrams.
Add a description, image, and links to the mechanical-sympathy topic page so that developers can more easily learn about it.
To associate your repository with the mechanical-sympathy topic, visit your repo's landing page and select "manage topics."