Anatomy of Linux Process Management

[ Thanks to An Anonymous Reader for this link. ]

"Linux is a very dynamic system with constantly changing computing needs. The representation of the computational needs of Linux centers around the common abstraction of the process. Processes can be short-lived (a command executed from the command line) or long-lived (a network service). For this reason, the general management of processes and their scheduling is very important.

"From user-space, processes are represented by process identifiers (PIDs). From the user's perspective, a PID is a numeric value that uniquely identifies the process. A PID doesn't change during the life of a process, but PIDs can be reused after a process dies, so it's not always ideal to cache them.

"In user-space, you can create processes in any of several ways. You can execute a program (which results in the creation of a new process) or, within a program, you can invoke a fork or exec system call. The fork call results in the creation of a child process, while an exec call replaces the current process context with the new program. I discuss each of these methods to understand how they work."

Complete Story

Related Stories:

• Using Kernel Customization Tools(Dec 23, 2008)
• Magic Tricks With the Sysreq Key(Dec 10, 2008)
• What is trackerd and Why Is It Running?(Dec 04, 2008)
• Creating a Daemon With Python(Nov 21, 2008)
• Getting CPU Information From Various Flavours Of Linux And Unix(Nov 14, 2008)
• Tip of the Trade: Proc and Processes(Nov 12, 2008)
• Ubuntu Multiple Terminals With Screen(Oct 07, 2008)