Ytdyklly

I have two OS on my pc. I installed Ubuntu 18.04 alongside with Windows 10. Total ram is 3.94 gb.

On Ubuntu my total ram is bigger than sum of free ram and used ram by 1,5 gb. And buff/cache is 1.5gb.

Is it normal, and it is just how linux works and if i will need more memory it will take from buff/cache? if not how to solve this problem?

$ watch -n 5 free -m

Every 5,0s: free -m                                                                                                     gleb: Sat Jul 21 14:34:49 2018



              total        used        free      shared  buff/cache   available

Mem:           3851        2048         214         343        1588        1184

Swap:          2047           0        2047





$ vmstat -s | head -10

gleb@gleb:~$ vmstat -s | head -10

  3943920 K total memory

  2102300 K used memory

  2350048 K active memory

   984768 K inactive memory

   209772 K free memory

    38824 K buffer memory

  1593024 K swap cache

  2097148 K total swap

        0 K used swap

  2097148 K free swap





$ cat /proc/meminfo | head -10

MemTotal:        3943920 kB

MemFree:          181792 kB

MemAvailable:    1126248 kB

Buffers:           32236 kB

Cached:          1463972 kB

SwapCached:            0 kB

Active:          2340464 kB

Inactive:        1021944 kB

Active(anon):    1691844 kB

Inactive(anon):   588964 kB



$ sudo lshw

description: Notebook

product: 80SM (LENOVO_MT_80SM_BU_idea_FM_Lenovo ideapad 310-15ISK)

vendor: LENOVO

version: Lenovo ideapad 310-15ISK

serial: PF0L15FM

width: 64 bits

...

-memory

      description: System Memory

      physical id: 26

      slot: System board or motherboard

      size: 4GiB

     -bank:0

         description: SODIMM DDR4 Synchronous 2133 MHz (0,5 ns)

         product: M471A5244BB0-CPB

         vendor: Samsung

         physical id: 0

         serial: 00000000

         slot: ChannelA-DIMM0

         size: 4GiB

         width: 64 bits

         clock: 2133MHz (0.5ns)

The way Linux handles the RAM depends on the Linux kernel version but in big picture, it goes roughly like this:

The actual hardware has X amount of RAM. The Linux boots and takes control of the RAM as much as possible (some may be reserved by BIOS/firmware, some other part by integrated graphics, etc). The rest of the RAM is first free.

During the kernel boot the Linux kernel locates the filesystem and starts init process (nowadays SysV init, Upstart or Systemd) that needs some RAM and will start additional processes from that point forward. All content that has been fethed from the filesystem will be kept in cache if possible. This allows kernel to skip reading the filesystem (and hard drive) in the future if the same data is needed again. The RAM that init or any other user mode process needs for processing instructions in CPU is called used.

By default, every time a user mode process writes something to the filesystem, the kernel will instead put the written contents to buffers and pretend to the user mode process that the data was already written to the filesystem. The user mode process continues and the kernel proceeds to write the buffers to the filesystem (and hard drive) in paraller.

In the end, used RAM is the part of the RAM that cannot be easily freed if another process needs more RAM. free RAM is part of the RAM that is not used for anything (practically wasted RAM) and buffer or cache are part of RAM that is used to speed up I/O. If some process needs more RAM than is free, the buffer and cache RAM is sacrificed but overall performance may suffer as a result.

As for the problem you mentioned in the comments about system working slower when some specific application is run, you would need to share more information about that specific case. Most probably the application you run on Linux and the "same" application on Windows are not the same thing.