Scheduled Maintenance: We are aware of an issue with Google, AOL, and Yahoo services as email providers which are blocking new registrations. We are trying to fix the issue and we have several internal and external support tickets in process to resolve the issue. Please see: viewtopic.php?t=158230

 

 

 

RAM comparison thread

Here you can discuss every aspect of Debian. Note: not for support requests!
Post Reply
Message
Author
User avatar
edbarx
Posts: 5401
Joined: 2007-07-18 06:19
Location: 35° 50 N, 14 º 35 E
Been thanked: 2 times

Re: RAM comparison thread

#16 Post by edbarx »

dzz wrote:Running TDE here on a number of Debian systems since it was first forked. I never experienced what edbarx wrote about TDE in dozens+ of TDE installs (but not nearly as good as Hallvor posted). It's normally similar resource usage to xfce4, which I also use. A TDE installation can be anything from tde-base-trinity (basic DE) to tde-trinity (monster meta). That might make a difference, I don't install the meta. Whether you use a display manager (on this box, TDM) or startx likely counts.
To be fair, I installed Trinity Desktop using tde-trinity which pulled many packages that I didn't want to install. The selection of automatically chosen software was clearly aimed at common computer users who only use a computer to chat, surf the net, write emails, watch videos, listen to music and share files.

If I remember well there was also file indexing going on, sudo was installed and su failed to recognise my root password.

ADDED LATER:
The following memory usage is after I stopped slim and XFCE using the command:

Code: Select all

# /etc/init.d/slim stop
It seems the kernel alone is using ~420 Mega Bytes! My installed kernel version is 3.2.0-4-rt-amd64. Running startx after stopping slim and saving the results shown below to a text file resulted into XFCE being loaded almost instantly. This shows that the kernel is doing some type of memory caching so that already used programs would use the same memory image avoiding reading and loading their executable code into memory. Unless there is a way of looking into what memory is chached by the kernel, memory use reporting alone is next to useless. Kernel memory caching is a brilliant idea to improve overall system response and to reduce having to read data from disks.

Code: Select all

             total       used       free     shared    buffers     cached
Mem:          1941        466       1474          6         41        327
-/+ buffers/cache:         97       1843
Swap:            0          0          0

 Private  +   Shared  =  RAM used	Program

116.0 KiB +  18.0 KiB = 134.0 KiB	uuidd
196.0 KiB +  34.5 KiB = 230.5 KiB	atd
232.0 KiB +  61.5 KiB = 293.5 KiB	init
188.0 KiB + 164.5 KiB = 352.5 KiB	avahi-autoipd (2)
328.0 KiB +  38.0 KiB = 366.0 KiB	rpc.idmapd
384.0 KiB +  38.5 KiB = 422.5 KiB	gpg-agent
324.0 KiB + 104.5 KiB = 428.5 KiB	cron
380.0 KiB +  57.0 KiB = 437.0 KiB	acpid
456.0 KiB + 153.5 KiB = 609.5 KiB	rpcbind
664.0 KiB + 162.5 KiB = 826.5 KiB	rpc.statd
512.0 KiB + 325.5 KiB = 837.5 KiB	su
812.0 KiB +  71.0 KiB = 883.0 KiB	dbus-daemon
680.0 KiB + 324.5 KiB =   1.0 MiB	login
740.0 KiB + 332.5 KiB =   1.0 MiB	getty (5)
  1.1 MiB +  86.5 KiB =   1.2 MiB	udevd
  1.3 MiB +  76.5 KiB =   1.3 MiB	rsyslogd
  1.5 MiB +  50.0 KiB =   1.5 MiB	wpa_supplicant
  1.3 MiB + 647.0 KiB =   1.9 MiB	polkitd
  1.6 MiB + 831.0 KiB =   2.4 MiB	console-kit-daemon
  2.2 MiB + 123.5 KiB =   2.4 MiB	privoxy
  2.2 MiB + 762.0 KiB =   3.0 MiB	upowerd
  3.0 MiB + 969.0 KiB =   3.9 MiB	bash (2)
 13.6 MiB + 183.0 KiB =  13.8 MiB	dhclient (2)
---------------------------------
                         39.1 MiB
=================================
Debian == { > 30, 000 packages }; Debian != systemd
The worst infection of all, is a false sense of security!
It is hard to get away from CLI tools.

alderaan
Posts: 90
Joined: 2013-07-25 20:20

Re: RAM comparison thread

#17 Post by alderaan »

I 'll try to make mine a little more scientific (lol).

Machine 1:
CPU: Intel Core 2 Duo E6550 @2.33 GHz [64-bit]
Video: Intel GMA 3100
RAM: 4 GB
OS: Debian 8 Jessie (Stable) [64-bit]
DE: XFCE
Applications:
Terminal: xfce-4 terminal running only top and one tab for the memory check
Web browser: Iceweasel (2 windows, 12 tabs total)
File manager: Thunar (2 tabs)
Text Editor: Mousepad


(A) Recovery mode (just after booting)
(B) CLI booting with the parameter: "systemd.unit=multi-user.target" on the kernel line
(C) GUI but zero programms running (just after booting)
(D) Busy: GUI but no uptime
(E) Busy and after 21 days uptime

Code: Select all

Session      free -m        ps_mem.py
  (A)          140           28.7 MiB
  (B)          187           58.0 MiB
  (C)          428          195.1 MiB
  (D)         1163          700.6 MiB
  (E)         3447            1.1 GiB
Pretty impressive, isn't it?
Some notes:
* Between sesions (E) and (D) the Iceweasel is dropping from 869 to 700 MiB (according to ps_mem.py) with exactly the same tabs opened!
* After the womping 869 MiB of Icewesel in session (E) the next biggest number is 24.3 MiB for xfce4-appfinder.

======================================================================================================================

Machine 2:

CPU: Intel Pentium 4 3.20 Ghz [64-bit]
Video: Integrated Graphics
RAM: 1.5 GiB
OS: Debian 8 Jessie (Stable) [64-bit]
DE: LXDE

Applications:

Terminal: LX Terminal running only top and one tab for the memory check
Web browser: Iceweasel (3 windows, 23 tabs total)
E-mail client: Icedove (many big mail accounts + Usenet + RSS)
File manager: PCManFM (2 windows, 5 tabs total)
Video player: VLC playing a 2.4 GiB movie (xvid.avi file)
Libre Office: 3 spreadsheat files (totaling size: 1.1 MiB)
Torrent Client: Transmission (Downloading 1 file and seeding 8)
Text Editor: Leafpad


(A) Recovery mode (just after booting)
(B) CLI booting with the parameter: "systemd.unit=multi-user.target" on the kernel line
(C) GUI but zero programms running (just after booting)
(D) Busy: GUI but no uptime
(E) Busy and after 6 days uptime

Code: Select all

Session       free -m       ps_mem.py
  (A)          99           20.7 MiB
  (B)         135           42.4 MiB
  (C)         368          158.1 MiB
  (D)         1408         961.3 MiB
  (E)         1396         718.7 MiB
How memory usage is bigger in (D) than (E), I can't explain!


The same machine with a quite minimal Sid installation that I use for experimenting and learning (with XFCE and Enlightenment available hence the two numbers in the (C) session):

Code: Select all

Session       free -m       ps_mem.py
  (A)          20             9.2 MiB
  (B)          35            42.0 MiB
  (C)         112/65        167/104 MiB
=> Enlightenment is quite lighter than XFCE

User avatar
None1975
df -h | participant
df -h | participant
Posts: 1404
Joined: 2015-11-29 18:23
Location: Russia, Kaliningrad
Has thanked: 46 times
Been thanked: 69 times

Re: RAM comparison thread

#18 Post by None1975 »

I installed Trinity Desktop R14.0.4, using tde-trinity metapackage. On cold startup, with default configuration, it uses only 200 mb. of RAM (x64 bit. machine). Very lightwieght (i think). :D
OS: Debian 12.4 Bookworm / DE: Enlightenment
Debian Wiki | DontBreakDebian, My config files on github

User avatar
NFT5
df -h | grep > 20TiB
df -h | grep > 20TiB
Posts: 598
Joined: 2014-10-10 11:38
Location: Canberra, Australia
Has thanked: 10 times
Been thanked: 43 times

Re: RAM comparison thread

#19 Post by NFT5 »

AMD FX6300 with 8GB RAM, MATE & Marco.

Code: Select all

chris@BOSSDESK:~$ free -m
             total       used       free     shared    buffers     cached
Mem:          7972        816       7155         18         40        316
-/+ buffers/cache:        460       7512
Swap:        15943          0      15943
chris@BOSSDESK:~$ free -m
             total       used       free     shared    buffers     cached
Mem:          7972       1724       6247         26        103        728
-/+ buffers/cache:        893       7079
Swap:        15943          0      15943
chris@BOSSDESK:~$ free -m
             total       used       free     shared    buffers     cached
Mem:          7972       4973       2998         32        120       1534
-/+ buffers/cache:       3318       4654
Swap:        15943          0      15943
chris@BOSSDESK:~$ 
First is on clean boot.

Second is with:
Firefox 45.5 with 2 tabs open
OpenOffice Writer with a 2 page document open
VLC playing an MP3
MATE Terminal
Dolphin file manager

Third is more like I usually run:
Firefox with 2 tabs open
MATE Terminal
Dolphin file manager
WinXP running in Virtualbox with a database open and 2GB RAM allocated to that VM
Thunderbird open

Chews up the RAM pretty quick and I've been using Win7 Pro in another VM which has 4GB allocated so really starting to think about upping RAM to 16GB. Swap is on an SSD so doesn't penalise me too much at present when it is used.

User avatar
Head_on_a_Stick
Posts: 14114
Joined: 2014-06-01 17:46
Location: London, England
Has thanked: 81 times
Been thanked: 133 times

Re: RAM comparison thread

#20 Post by Head_on_a_Stick »

wizard10000 wrote:slim
That project is dead upstream and the display manager doesn't handle systemd user sessions correctly, this can cause problems (if you use systemd).

I would recommend LightDM instead.

Or https://bbs.archlinux.org/viewtopic.php ... 6#p1680406 :D
deadbang

Post Reply