When every microsecond counts, inter-thread communication must be lean and predictable. This talk dives into the design of a high-performance SPSC bounded queue for ultra-low latency trading systems. We’ll cover eliminating locks with atomics, reducing sync costs with memory ordering, and avoiding traps like false sharing, all backed by real-world measurements and cache analysis. Attendees will gain practical strategies for writing lock-free data structures that deliver under pressure.
