# Identifying system performance bottlenecks # preface,
1. Monitoring:
2. test:
3. Optimization:
All common linux Performance Tools, including monitoring, testing, and optimization, are shown in the figure Brendan D. Gregg (Linux Performance Tools)
I want to briefly introduce all the tools here, but I want to use a few simple examples instead of a large and complete reference document, introduce the common usage and scenarios of these tools in practical ways, so that you can quickly get started with solving system performance problems.
Why do we need to learn so many tools? I think learning more than one tool will give you one more idea, and more skills will not be overwhelmed. Especially when you encounter a difficult performance problem, a sudden inspiration will help you solve the problem instantly.
There are dozens of tools in this batch. I will start from monitoring and slowly introduce these tools all over again. However, the order of introduction is not necessarily as follows, in addition, I will introduce associated tools together.
1 monitoring tool: Application Layer
Ltrace perf
System Call Interface
Strace sysdig perf
File System,
Pidstat iostat
Network,
Ss netstat iptraf tcpdump
Scheduling and virtual memory
Perf mpstat top ps pidstat vmstat slabtop free
Device drive
IO Bus Device
Perf tiptop
IO controller,
Iostat iotop blktrace swapon
Network Controller
Iptraf ethtool snmpget lldptool nicstat netstat ip
Others
Sar dstat/proc
Perf ftrace stap ktap ebpf dtrace lttng
2 test tool :...
3. Optimization Tool :...
================ The following is the body of the text ================================#locate system performance bottleneck # strace & ltrace
Copyright Disclaimer: This article is an original article by the blogger and cannot be reproduced without the permission of the blogger.