Unix / Linux – System Performance

In this section we learnt about Unix / Linux – System Performance. We will introduce you to a few free tools that are available to monitor and manage performance on Unix systems. In these digonise we know how to fix unix systems performance.

Unix has different major resource types that need to be monitored and tuned −

  • CPU
  • Memory
  • Disk space
  • Communications lines
  • I/O Time
  • Network Time
  • Applications programs

Performance Components

The table lists out five major components-

Sr.No.Component & Description
1User State CPUThe actual amount of time the CPU spends running the users’ program in the user state. It includes the time spent executing library calls, but does not include the time spent in the kernel on its behalf
2System State CPUThis is the amount of time the CPU spends in the system state on behalf of this program. All I/O routines require kernel services. The programmer can affect this value by blocking I/O transfers
3I/O Time and Network TimeThis is the amount of time spent moving data and servicing I/O requests
4Virtual Memory PerformanceThis includes context switching and swapping
5Application ProgramTime spent running other programs – when the system is not servicing this application because another application currently has the CPU

Performance Tools

Unix provides important tools to measure unix system performance-

Sr.No.Command & Description
1nice/reniceRuns a program with modified scheduling priority
2netstatPrints network connections, routing tables, interface statistics, masquerade connections, and multicast memberships
3timeHelps time a simple command or give resource usage
4uptimeThis is System Load Average
5psReports a snapshot of the current processes
6vmstatReports virtual memory statistics
7gprofDisplays call graph profile data
8profFacilitates Process Profiling
9topDisplays system tasks

In this section we learnt about Unix / Linux – System Performance.

To know more Click Here

This Post Has 2 Comments

Leave a Reply