Signals and Threads Performance Engineering on Hard Mode with Andrew Hunter
45 snips
Nov 28, 2023 Andrew Hunter, a performance engineer who makes code really fast, discusses optimizing systems at hyperscale and the challenges of optimizing trading systems. He shares his favorite profiling techniques, tools for visualizing traces, and the unique challenges of optimizing OCaml versus C++. They also touch on the joys of musical theater and how to pass an interview when sleep-deprived.
AI Snips
Chapters
Transcript
Episode notes
Andrew's Origin in Performance
- Andrew Hunter discovered his passion for performance engineering through deep curiosity about system internals during college and grad school.
- He devoted himself to understanding detailed operating system and hardware mechanics, which naturally led to focusing on performance aspects.
Scale Enables Easy Mode
- Large companies find performance gains easier because any small optimization scales from huge usage.
- Common infrastructure benefits all business logic, creating a "data center tax" but also huge leverage for improvements.
Challenging Low-Latency Trade-offs
- Without hyperscale, micro-optimizations have little financial impact and most time is spent waiting idly in trading systems.
- Trading systems prioritize low latency during bursts, requiring focus on latency over utilization or throughput.