1. Overview
free and top are built-in Linux commands used to get detailed reports on a system’s memory usage. These commands are often misinterpreted by users because neither is very clear about what it actually measures.
In this tutorial, we’ll discuss these commands and look at some of the similarities and differences between them.
2. free
free is a popular and powerful Unix command that gives information about memory usage in a human-readable format. It shows the total amount of free and used memory on the system. It also includes physical space, swap spaces, buffers, and caches used by the kernel. By default, it displays memory values in kilobytes(kb).
The information is retrieved by parsing the /proc/meminfo file. We can view it using the cat command:
$ cat /proc/meminfo
MemTotal: 8007956 kB
MemFree: 195700 kB
MemAvailable: 2721768 kB
Buffers: 420764 kB
...truncated...
The free command has this basic syntax:
$ free [options]
It also has a lot of options that we can use to manipulate the display of data. Let’s use the -h to display the output in human-readable format:
$ free -h
total used free shared buff/cache available
Mem: 7.6Gi 4.3Gi 264Mi 623Mi 3.1Gi 2.5Gi
Swap: 2.0Gi 6.0Mi 2.0Gi
Here’s what each of these columns represents:
- total: refers to the total amount of memory available for use by running applications and services
- used: shows the amount of used memory. It’s calculated as used = total – free – buffers – cache
- free: shows the amount of unused memory
- shared: shows memory used mostly by tmpfs
- buff/cache: shows the combined memory usage from kernel buffers, page cache, and slabs
- available: shows an estimate of the amount of memory available for starting new applications without swapping
We can also continuously display memory information on the terminal by using the -s (–seconds) option:
$ free -s 2
total used free shared buff/cache available
Mem: 8007956 4711664 224108 638984 3072184 2367976
Swap: 2097148 13324 2083824
total used free shared buff/cache available
Mem: 8007956 4720488 219068 635188 3068400 2362948
Swap: 2097148 13324 2083824
The command above displays memory usage information every 2 seconds. We can notice a difference in some of the values between the two outputs because the values update regularly.
3. top
The top (table of processes) command allows us to monitor in real-time the number of processes running and kernel-managed tasks on a Linux machine. It also offers us a system information summary that displays resource usage, inclusive of CPU and memory usage.
Its interface is split into two parts. The top part displays the stats values, while the lower part displays the list of the running processes. It also supports color, highlighting, and even elementary graphs. Furthermore, it’s interactive, which allows us to browse through the list of processes and even kill a process.
There are a lot of versions of the top command. Let’s view our current version:
$ top -v
procps-ng 3.3.16
Usage:
top -hv | -bcEHiOSs1 -d secs -n max -u|U user -p pid(s) -o field -w [cols]
In this tutorial, we’ll be working with v3.3.16 which comes with the procps-ng package.
By default, the top command produces this output:
$ top
...
%Cpu(s): 7.2 us, 2.3 sy, 0.0 ni, 90.5 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
MiB Mem : 7820.3 total, 239.9 free, 6660.3 used, 920.0 buff/cache
MiB Swap: 2048.0 total, 1571.5 free, 476.5 used. 450.8 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
2832 user 20 0 4285156 194344 34556 S 6.0 2.4 22:29.39 gnome-sh+
...
We can even display more columns. To do this, we need to press the letter f on the top command’s dashboard:
...
Navigate with Up/Dn, Right selects for move then <Enter> or Left commits,
'd' or <Space> toggles display, 's' sets sort. Use 'q' or <Esc> to end!
* PID = Process Id GID = Group Id SUPGIDS = Supp Groups IDs RSlk = RES Locked (KiB)
* USER = Effective User Name GROUP = Group Name SUPGRPS = Supp Groups Names RSsh = RES Shared (KiB)
* PR = Priority PGRP = Process Group Id TGID = Thread Group Id CGNAME = Control Group name
...
Next, let’s navigate to CODE, DATA, and SWAP columns, activate them with the space button and then press the Esc button to return to the main dashboard.
We should have this output:
...
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND SWAP CODE DATA
15493 user 20 0 32.4g 224864 88356 S 51.3 2.8 48:48.97 chrome 0 176216 332228
5346 user 20 0 28.4g 338256 95904 S 46.4 4.2 36:07.86 chrome 0 176216 521552
...
Let’s understand more about what most of these columns represents:
- %MEM: shows the percentage use of the total physical memory by a process
- VIRT: show total memory that a process has access to, including shared memory, swapped pages, and mapped pages
- RES: shows the total physical memory used, including private and shared by a process
- SHR: shows total physical shared memory used by a process
- DATA: shows the total private memory used by a process, both physical and virtual
- CODE: shows the total physical memory utilized to load applications
- SWAP: shows the total amount of swap memory available
4. Comparing the Output of free and top Commands
Let’s compare the results of the free and top commands and see some of the similarities and differences that exist between them.
First, let’s display memory usage information using free in megabytes (MB):
$ free -m
total used free shared buff/cache available
Mem: 7820 5887 890 513 1042 1151
Swap: 2047 704 1343
Next, let’s display the information from the top command. Display of the values is in megabytes by default:
$ top
top - 15:57:32 up 9:17, 1 user, load average: 2.40, 2.07, 1.63
Tasks: 401 total, 1 running, 400 sleeping, 0 stopped, 0 zombie
%Cpu(s): 14.5 us, 1.9 sy, 0.0 ni, 83.5 id, 0.1 wa, 0.0 hi, 0.0 si, 0.0 st
MiB Mem : 7820.3 total, 817.9 free, 5957.3 used, 1045.1 buff/cache
MiB Swap: 2048.0 total, 1343.5 free, 704.5 used. 1092.0 avail Mem
...
The first thing we can notice is that both commands have almost similar values on:
- the total amount of memory available, used, and free
- the total amount of swap memory available, used, and free
- buff/cache memory size
The slight difference in some of the values is because of the time delay between running the two commands.
We can also notice that the total memory is 7820 MB, with 5887 MB used and 890MB free. However, that doesn’t mean that running applications can only request the 890MB free memory. The 1042MB allocated for buffers and cache can be freed up to yield more memory for new application requests.
Linux tries to use RAM to optimize disk operations by making use of available memory for creating buffers and cache. This reduces I/O operations and helps the system run faster.
5. Conclusion
In this article, we’ve discussed the top and free commands that are used to retrieve reports on a system’s memory usage. We also looked at some of the differences and similarities that exist between these commands and their respective outputs.