trading systems

A 5.7 MB binary wire block stream took 73 milliseconds to parse with standard MessagePack deserialization, stalling the ingest thread during market bursts. How an AVX-512 scanner and fixed-point integers dropped per-block latency from 1,623 µs to 129 µs and brought throughput to 978 MB/s.

A market-data ring in /dev/shm now has identical copies on other hosts, same sequence numbers, same order: 30 µs across a LAN by multicast, 51.7 ms p99 across the Pacific.

22 exchanges (Binance, OKX, Bybit, Coinbase and more), 240,000 records per second, zero crashes all summer, and a dozen incidents where data vanished silently. Where to put the collector, what breaks without an error in the log, and which metrics give it away.

The hand-off between a market-data reader and a strategy loop is one of the few latencies in a trading process under our control. How we built a lock-free channel for it, what broke on a 128-core ARM box, and where the assembly landed.