top is an extremely useful program that acts similar to Windows Task Manager or MacOS's Activity Monitor. Running top on your *nix machine (*nix means an operating system that behaves in a manner similar to that of a UNIX operating system) will show you a live, running view of the process running on your system.
$ top
Depending on which version of top you're running, you'll get something that looks like this:
top - 08:31:32 up 1 day, 4:09, 0 users, load average: 0.20, 0.12, 0.10
Tasks: 3 total, 1 running, 2 sleeping, 0 stopped, 0 zombie
%Cpu(s): 0.5 us, 0.3 sy, 0.0 ni, 99.2 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
KiB Mem: 4042284 total, 2523744 used, 1518540 free, 263776 buffers
KiB Swap: 1048572 total, 0 used, 1048572 free. 1804264 cached Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
1 root 20 0 21964 3632 3124 S 0.0 0.1 0:00.23 bash
193 root 20 0 123520 29636 8640 S 0.0 0.7 0:00.58 flask
195 root 20 0 23608 2724 2400 R 0.0 0.1 0:00.21 top
Your version of top may look different from this, particularly in the columns that are displayed.
How to read the output
You can tell what you're running based on the output, but trying to interpret the results can be slightly confusing.
The first few lines contain a bunch of statistics (the details) followed by a table with a list of results (the list). Let's start with the latter.
The list
These are the processes that are running on the system. By default, they are ordered by CPU usage in descending order. This means the items at the top of the list are using more CPU resources and causing more load on your system. They are literally the "top" processes by resource usage. You have to admit, it's a clever name.
The COMMAND column on the far right reports the name of the process (the command you ran to start them). In this example, they are bash (a command interpreter we're running top in), flask (a web micro-framework written in Python), and top itself.
The other columns provide useful information about the processes:
• PID: the process id, a unique identifier for addressing the processes
• USER: the user running the process
• PR: the task's priority
• NI: a nicer representation of the priority
• VIRT: virtual memory size in KiB (kibibytes)*
• RES: resident memory size in KiB* (the "physical memory" and a subset of VIRT)
• SHR: shared memory size in KiB* (the "shared memory" and a subset of VIRT)
• S: process state, usually I=idle, R=running, S=sleeping, Z=zombie, T or t=stopped (there are also other, less common options)
• %CPU: Percentage of CPU usage since the last screen update
• %MEM: percentage of RES memory usage since the last screen update
• TIME+: total CPU time used since the process started
• COMMAND: the command, as described above
Knowing exactly what the VIRT, RES, and SHR values represent doesn't really matter in everyday operations. The important thing to know is that the process with the most VIRT is the process using the most memory. If you're in top because you're debugging why your computer feels like it's in a pool of molasses, the process with the largest VIRT number is the culprit. If you want to learn exactly what "shared" and "physical" memory mean, check out "Linux Memory Types" in the top manual.
* And, yes, I did mean to type kibibytes, not kilobytes. The 1,024 value that you normally call a kilobyte is actually a kibibyte. The Greek kilo means thousand and means 1,000 of something (e.g., a kilometer is a thousand meters, a kilogram is a thousand grams). Kibi is a portmanteau (a word blending the sounds and combining the meanings of two others) of kilo and byte, and it means 1,024 bytes (or 210). But, because words are hard to say, many people say kilobyte when they mean 1,024 bytes. All this means is top is trying to use the proper terms here, so just go with it.
A note on screen updates:
Live screen updates are one of the objectively really cool things Linux programs can do. This means they can update their own display in real time, so they appear animated. Even though they're using text. So cool! In our case, the time between updates is important, because some of our statistics (%CPU and %MEM) are based on the value since the last screen update.
And because we're running in a persistent application, we can press key commands to make live changes to settings or configurations (instead of, say, closing the application and running the application again with a different command-line flag).
Typing h invokes the "help" screen, which also shows the default delay (the time between screen updates). By default, this value is (around) three seconds, but you can change it by typing d (presumably for "delay") or s (probably for "screen" or "seconds").
The details
Above the list of processes, there's a whole bunch of other useful information. Some of these details may look strange and confusing, but once you take some time to step through each one, you'll see they're very useful stats to pull up in a pinch.
The first row contains general system information
• top: we're running top! Hi top!
• XX:YY:XX: the time, updated every time the screen updates
up (then X day, YY:ZZ): the system's uptime, or how much time has passed since the system turned on
• load average (then three numbers): the system load over the last one, five, and 15 minutes, respectively
The second row Tasks shows information about the running tasks, and it's fairly self-explanatory. It shows the total number of processes and the number of running, sleeping, stopped, and zombie processes. This is literally a sum of the S (state) column described above.
The third row %Cpu(s) shows the CPU usage separated by types. The data are the values between screen refreshes. The values are:
• us: user processes
• sy: system processes
• ni: nice user processes
• id: the CPU's idle time; a high idle time means there's not a lot going on otherwise
• wa: wait time, or time spent waiting for I/O completion
• hi: time spent waiting for hardware interrupts
• si: time spent waiting for software interrupts
• st: "time stolen from this VM by the hypervisor"
You can collapse the Tasks and %Cpu(s) rows by typing t (for "toggle").
The fourth (KiB Mem) and fifth rows (KiB Swap) provide information for memory and swap. These values are:
•total
•used
•free
But also:
• memory buffers
• swap cached Mem
By default, they're listed in KiB, but pressing E (for "extend memory scaling") cycles through different values: kibibytes, mebibytes, gibibytes, tebibytes, pebibytes, and exbibytes. (That is, kilobytes, megabytes, gigabytes, terabytes, petabytes, and exabytes, but their "real names.")
The top user manual shows even more information about useful flags and configurations. To find the manual on your system, you can run man top. There are various websites that show an HTML rendering of the manual, but note that these may be for a different version of top.
Katie McLaughlin has worn many different hats over the years. She has previously been a software developer for many languages, systems administrator for multiple operating systems, and speaker on many different topics. When she's not changing the world, she enjoys making cooking, tapestries, and seeing just how well various application stacks handle emoji.
Opensource.com aspires to publish all content under a Creative Commons license but may not be able to do so in all cases. You are responsible for ensuring that you have the necessary permission to reuse any work on this site. Red Hat and the Red Hat logo are trademarks of Red Hat, Inc., registered in the United States and other countries. This work is licensed under a Creative Commons Attribution-Share Alike 4.0 International License.
More Windows Administration Information:
• What is Linux?
• The Linux Directory Structure
• 7 Steps to Securing Your Linux Server
• What is a Linux Distribution
• Easy Way to Install Linux
• Understanding Linux Filesystems: Ext4 and Beyond
• View Your Linux Server's Network Connections With Netstat
• How to Switch to Ubuntu
• Levels and Layers of Abstraction in a Linux System
• Linux Memory User Space