0 views collected around this technical thread.
This article investigates two real‑world performance incidents—one caused by excessive disk I/O from a misconfigured Filebeat and another by kernel memory‑fragmentation bugs triggered by a trace feature—detailing observations, Linux diagnostic commands, analysis, and practical remediation steps.
This article introduces Linux perf, explains how to install it, demonstrates basic commands such as perf‑list, perf‑stat, perf‑top, perf‑record and perf‑report, and shows how to combine perf with flame‑graphs to locate CPU‑bound hotspots and other performance bottlenecks in applications and the kernel.
This article explains how software performance events in the Linux kernel affect application performance, demonstrates how to list, count, and trace these events with perf, and shows how to visualize the results using FlameGraph to pinpoint bottlenecks such as cache misses, context switches, and page faults.
This article explains how to generate and interpret flame graphs for CPU performance profiling on Linux, covering the use of perf for sampling, the underlying kernel mechanisms, and the processing steps with Brendan Gregg's FlameGraph scripts to visualize hot functions.
This article explains the concepts and mechanisms behind Linux kernel tracing tools—including ftrace, perf, kprobes, uprobes, tracepoints, ftrace, perf events, and eBPF—showing how probes are injected, how trace data is collected, and which technology to choose for different debugging and performance scenarios.
Using eBPF to bypass traditional bridge, netfilter, and routing steps, this article demonstrates how container network packet forwarding can be accelerated, reducing receive and transmit latency by around 40‑82%, with detailed perf‑ftrace analysis, code examples, and performance comparisons of three forwarding modes.
This comprehensive guide explains Linux kernel performance profiling—both on‑CPU and off‑CPU—by stressing the need to target the critical 3 % of code, covering throughput, latency and power metrics, scalability laws, flame‑graph visualizations, perf and eBPF tools, lock‑contention analysis, and further reading recommendations.
The article explains how a MySQL 5.6 master‑slave setup experienced growing replication lag on an off‑site replica, how the issue was traced to page‑compression on a large table using show slave status, perf and pstack, and how disabling compression resolved the delay.
This article provides a comprehensive guide to game pressure testing, covering the full workflow from requirement communication and test planning to execution, performance analysis using tools like LuaProfiler, Perf, and VTune, and detailed result interpretation with case studies.
In large Redis clusters the original CLUSTER SLOTS implementation traversed every master and all 16,384 slots, causing ~52 % CPU usage and high MGET latency during migrations, but redesigning it to iterate the pre‑built slot array reduced complexity to O(total slots), cutting execution time from 2 ms to 0.17 ms and eliminating the CPU hotspot, a fix now merged into Redis 6.2.2.
This article explains how to diagnose a sudden rise in MySQL file handle usage by reproducing the issue with large tables, using Linux perf to trace open system calls, analyzing hash join behavior, and querying performance_schema.file_instances to pinpoint the underlying cause.
This tutorial demonstrates how to reproduce a MySQL memory‑spike bug, monitor the process with Linux perf to capture mmap system calls, and analyze the resulting perf.out to identify which SQL statements trigger large memory allocations, while also discussing the method’s limitations.
This article explains how Linux tracing tools such as kprobes, uprobes, tracepoints, eBPF, perf, ftrace, sysdig, SystemTap and LTTng fit together, describing their data sources, collection mechanisms and front‑ends, and offers guidance on choosing the right tool for low‑overhead debugging and performance analysis.
This article explains how to install perf and FlameGraph tools on Linux, capture profiling data from MySQL using various perf record options, convert the binary output to a readable script, and generate an SVG flame graph for visual performance analysis, including example commands and interpretation tips.
This article explains how Linux reports CPU usage with tools like top, the meaning of the fields in /proc/stat, how utilization percentages are calculated, the concepts of run queues, load average, context switching, multi‑core scheduling, and how to use perf and taskset for deeper performance analysis.
This article documents a multi‑stage investigation of intermittent GitLab Runner build timeouts and hangs, describing the background, VM configuration, successive diagnostic steps using strace, iotop, perf, and storage‑driver adjustments, and concludes with performance test results and lessons learned.
This article explores practical Python debugging methods—from simple print statements and structured logging to powerful IDE debuggers, GDB, perf, strace, and the inject debugger pylane—providing code examples, best‑practice guidelines, and tips for improving observability and performance in production systems.
This article explains why the traditional %CPU metric can be deceptive, describes how CPU stalls and memory bottlenecks affect perceived utilization, and offers practical guidance on using IPC and performance counters to obtain a more accurate view of system performance.
This article introduces three popular Linux performance profiling tools—Perf, gprof, and Valgrind—explaining their installation, basic command‑line usage, and how to visualize results with flame graphs, call graphs, or KCachegrind, while comparing their intrusiveness, startup methods, and output formats.
This article explains how to detect whether inbound or outbound packets are being dropped on a Linux system, shows how to use netstat, dropwatch, and perf to pinpoint the cause, and provides step‑by‑step commands and example outputs for debugging packet loss.