Profit or Poverty: False Sharing
An exploration of false sharing in CPU caches, how it silently degrades performance, and why it matters in low-latency environments.
See all tags
6 posts in total
An exploration of false sharing in CPU caches, how it silently degrades performance, and why it matters in low-latency environments.
How High-Frequency Trading uses kernel bypass to sidestep the Linux network stack and achieve nanosecond latency.
Why NTP isn’t accurate enough for trading, and how PTP (Precision Time Protocol) achieves sub-microsecond precision.
Understanding virtual memory, the TLB (Translation Lookaside Buffer), and its performance impact on low-latency systems.
An overview of the Linux scheduler’s limitations and why realtime kernel patches are essential for time-critical applications.
An introduction to NUMA (Non-Uniform Memory Access) and why mastering it is a hard requirement for HFT micro-optimizations.