If you haven\u2019t been paying close attention, you might not have noticed a small but significant change in how Linux systems work with respect to runtime data. A re-arrangement of how and where it\u2019s accessible in the file system started taking hold about eight years ago. And while this change might not have been big enough of a splash to wet your socks, it provides some additional consistency in the Linux file system and is worthy of some exploration.\nTo get started, cd your way over to \/run. If you use df to check it out, you\u2019ll see something like this:\n$ df -k .\nFilesystem 1K-blocks Used Available Use% Mounted on\ntmpfs 609984 2604 607380 1% \/run\n\nIdentified as a \u201ctmpfs\u201d (temporary file system), we know that the files and directories in \/run are not stored on disk but only in volatile memory. They represent data kept in memory (or disk-based swap) that takes on the appearance of a mounted file system to allow it to be more accessible and easier to manage.\n\n\/run is home to a wide assortment of data. For example, if you take a look at \/run\/user, you will notice a group of directories with numeric names.\n$ ls \/run\/user\n1000 1002 121\n\nA long file listing will clarify the significance of these numbers.\n$ ls -l\ntotal 0\ndrwx------ 5 shs shs 120 Jun 16 12:44 1000\ndrwx------ 5 dory dory 120 Jun 16 16:14 1002\ndrwx------ 8 gdm gdm 220 Jun 14 12:18 121\n\nThis allows us to see that each directory is related to a user who is currently logged in or to the display manager, gdm. The numbers represent their UIDs. The content of each of these directories are files that are used by running processes.\nThe \/run\/user files represent only a very small portion of what you\u2019ll find in \/run. There are lots of other files, as well. A handful contain the process IDs for various system processes.\n$ ls *.pid\nacpid.pid atopacctd.pid crond.pid rsyslogd.pid\natd.pid atop.pid gdm3.pid sshd.pid\n\nAs shown below, that sshd.pid file listed above contains the process ID for the ssh daemon (sshd).\n$ cat sshd.pid\n1148\n$ ps -ef | grep sshd\nroot 1148 1 0 Jun14 ? 00:00:00 \/usr\/sbin\/sshd -D \/run\/lock\ndrwxrwxr-x 17 root syslog 4096 Jun 17 00:00 log\ndrwxrwsrwt 2 root mail 4096 Jun 13 12:10 mail\ndrwxrwsrwt 2 root whoopsie 4096 Jan 5 2018 metrics\ndrwxr-xr-x 2 root root 4096 Jan 5 2018 opt\nlrwxrwxrwx 1 root root 4 May 14 2018 run -> \/run\ndrwxr-xr-x 9 root root 4096 Jun 16 2018 snap\ndrwxr-xr-x 9 root root 4096 Jun 9 15:14 spool\ndrwxrwxrwt 8 root root 4096 Jun 17 00:00 tmp\ndrwxr-xr-x 3 root root 4096 Jan 19 12:14 www\n\nWhile minor as far as technical changes go, the transition to using \/run simply allows for a better organization of runtime data in the Linux file system.