- This topic has 20 replies, 12 voices, and was last updated Apr 26-3:47 pm by Robin.
-
Posts
-
mikey777
Everyone here is aware that antiX doesn’t have SystemD – but what are the advantages and disadvantages of not having it?
I’ve noticed when comparing x64 editions, antiX19.4 consumes half the amount of RAM (when idling) than Debian10: e.g. idling at ~175MB for antiXcore-LXDE, but at ~360M for Debian-LXDE.Is the above observed association of lower RAM consumption with the absence of SystemD causally related? Or is it due to something else?
Bodhi 6.0 (Moksha desktop, Ubuntu-based), when installed alongside antiX & Debian on my 14-year-old laptop, idles at only 220MB RAM, even though it comes prepackaged with SystemD. As in life, I suspect it’s down to a combination of factors resulting in a lower RAM in antiX or Bodhi.I’ll probably now get replies that go above my head, but I’m curious nevertheless to hear all your comments ..!
- This topic was modified 1 year, 9 months ago by mikey777.
▪ 32-bit antix19.4-core+LXDE installed on :
- (2011) Samsung NP-N145 Plus (JP04UK) – single-core CPU Intel Atom N455@1.66GHz, 2GB RAM, integrated graphics.
▪ 64-bit antix21-base+LXDE installed on:
- (2008) Asus X71Q (7SC002) – dual CPU Intel T3200@2.0GHz, 4GB RAM. Graphics: Intel Mobile 4 Series, integrated graphics
- (2007) Packard Bell Easynote MX37 (ALP-Ajax C3) – dual CPU Intel T2310@1.46GHz, 2GB RAM. Graphics: Silicon Integrated Systems.
sybok
Hi,
I doubt this difference in RAM is due to missing ‘systemd’ (only).
antiX has a lot of custom programs (under the hood) to minimize memory footprint.Since your post indicates that you have means of comparison (live USB, perhaps), examine memory consumption e.g. using top/htop, consider making screenshots and compare what is running.
You may wish to include other non-systemd distributions in your comparison, they can be searched at distrowatch.com, see this link.
You could try Debian + XFCE vs. Devuan + XFCE which could be a closest candidate to judge from simple systemd vs. other init without too many other changes.- This reply was modified 1 year, 9 months ago by sybok. Reason: explicit statement of commenting on RAM
olsztyn
Since your post indicates that you have means of comparison (live USB, perhaps), examine memory consumption e.g. using top/htop, consider making screenshots and compare what is running.
I happen to be of the same opinion as described in sybok’s post…
As related to the quoted part:
– Comparison test also should be run on the same hardware. Exactly the same antiX composition (Live or not) has lower memory footprint on lower specs machine than on higher specs one. The same Live antiX run on my Thinkpad X61 shows significantly lower footprint than when run on my Thinkpad T410, which is still lower than run on my Thinkpad T520, although the installed memory size on X61 is the same as on T410 (4GB). Comparison of footprint to T520 in this case might not be fair as T520 has 8GB memory.
– My understanding also is in line with sybok’s that systemd has probably little impact on memory footprint. I would be surprised if antiX owner chose no-systemd architecture for this reason, but it is not up to me to say…
I did a quick comparison myself and have not found a significant memory footprint difference on the same machine. But my comparison was not scientific, so irrelevant to make any conclusion.I think systemd got lots of bad feelings due to unjustified complexity, which is not consistent with Unix philosophy. Not only more difficult but also due to complexity prone to bugs. However, even more importantly, as dependency on systemd in Linux world became prevalent it came to the point that systemd virtually holds Linux hostage. A few distros chose to revolt against systemd, but considering little effort in development of traditional init systems (such as runit or S6) or new ones, this seems a battle up the hill and soon all Linux will fall victim of systemd monopoly…
This is for the above reason that if I have a choice I would rather run a simpler non-systemd init system and stand up against trend limiting freedom.
Just MHO…- This reply was modified 1 year, 9 months ago by olsztyn.
Live antiX Boot Options (Previously posted by Xecure):
https://antixlinuxfan.miraheze.org/wiki/Table_of_antiX_Boot_Parameters
christophe
… if I have a choice I would rather run a simpler non-systemd init system and stand up against trend limiting freedom.
Just MHO…My opinion, as well. 🙂
confirmed antiX frugaler, since 2019
mikey777
Hi,
Debian + XFCE vs. Devuan + XFCE which could be a closest candidate to judge from simple systemd vs. other init without too many other changes.Yes, I had thought of this but Devuan doesn’t seem to want to install on my multiboot system & don’t understand why. I tried both the USB-stick & CD route for installation, but both failed. Maybe a corrupt iso file. I might try again.
▪ 32-bit antix19.4-core+LXDE installed on :
- (2011) Samsung NP-N145 Plus (JP04UK) – single-core CPU Intel Atom N455@1.66GHz, 2GB RAM, integrated graphics.
▪ 64-bit antix21-base+LXDE installed on:
- (2008) Asus X71Q (7SC002) – dual CPU Intel T3200@2.0GHz, 4GB RAM. Graphics: Intel Mobile 4 Series, integrated graphics
- (2007) Packard Bell Easynote MX37 (ALP-Ajax C3) – dual CPU Intel T2310@1.46GHz, 2GB RAM. Graphics: Silicon Integrated Systems.mikey777
This is for the above reason that if I have a choice I would rather run a simpler non-systemd init system and stand up against trend limiting freedom.
Just MHO…MHO too 🙂
▪ 32-bit antix19.4-core+LXDE installed on :
- (2011) Samsung NP-N145 Plus (JP04UK) – single-core CPU Intel Atom N455@1.66GHz, 2GB RAM, integrated graphics.
▪ 64-bit antix21-base+LXDE installed on:
- (2008) Asus X71Q (7SC002) – dual CPU Intel T3200@2.0GHz, 4GB RAM. Graphics: Intel Mobile 4 Series, integrated graphics
- (2007) Packard Bell Easynote MX37 (ALP-Ajax C3) – dual CPU Intel T2310@1.46GHz, 2GB RAM. Graphics: Silicon Integrated Systems.mikey777
– Comparison test also should be run on the same hardware. Exactly the same antiX composition (Live or not) has lower memory footprint on lower specs machine than on higher specs one.
I always run OS-comparisons on the same laptop, as a multibooted system.
The same Live antiX run on my Thinkpad X61 shows significantly lower footprint than when run on my Thinkpad T410, which is still lower than run on my Thinkpad T520, although the installed memory size on X61 is the same as on T410 (4GB). Comparison of footprint to T520 in this case might not be fair as T520 has 8GB memory.
Yes, it’s an interesting observation. I’ve found much the same with my hardware – the more powerful the machine the greater the OS footprint. For example, on dual-core CPU machines, I’ve observed that antix19.4core-LXDE idles at 175MB RAM with a 2GHz processor (Asus X71Q), but only 95MB with a 1.4GHz processor (Packard Bell Easynote MX37). I would imagine this relationship between OS-footprint & machine power is a universal one …
- This reply was modified 1 year, 9 months ago by mikey777.
- This reply was modified 1 year, 9 months ago by mikey777.
- This reply was modified 1 year, 9 months ago by mikey777.
▪ 32-bit antix19.4-core+LXDE installed on :
- (2011) Samsung NP-N145 Plus (JP04UK) – single-core CPU Intel Atom N455@1.66GHz, 2GB RAM, integrated graphics.
▪ 64-bit antix21-base+LXDE installed on:
- (2008) Asus X71Q (7SC002) – dual CPU Intel T3200@2.0GHz, 4GB RAM. Graphics: Intel Mobile 4 Series, integrated graphics
- (2007) Packard Bell Easynote MX37 (ALP-Ajax C3) – dual CPU Intel T2310@1.46GHz, 2GB RAM. Graphics: Silicon Integrated Systems.sybok
Also, one could try MX Linux since it did/does support two init systems, systemd and sysvinit (?), I guess.
How could I have forgotten about this one…Regarding the memory tests, it seems possible to run
top > ~/top_test.txt
followed by ‘Ctrl+C’ to quit the top.
The file ‘~/top_test.txt’ contains some information from top but it is a partially garbled harder to read binary file.- This reply was modified 1 year, 9 months ago by sybok. Reason: Redirect top
- This reply was modified 1 year, 9 months ago by sybok.
mikey777
Also, one could try MX Linux since it did/does support two init systems, systemd and sysvinit (?), I guess.
How could I have forgotten about this one…Regarding the memory tests, it seems possible to run
top > ~/top_test.txt
followed by ‘Ctrl+C’ to quit the top.
The file ‘~/top_test.txt’ contains some information from top but it is a partially garbled harder to read binary file.Thanks syborg. Your idea of using MX Linux as a model system, for looking at differences in systemD vs. non-systemD, appears a really good one. I had a look at top, the first time I do this – it seems to provide a single snapshot measurement of htop. As you said, the text file is hard to read, giving some garbled characters on it!
Top seems a useful app, but I don’t really want to go into such detail. I was only interested in the observation that there appeared to be a difference in overall RAM consumption (shown by htop) between systemD vs non-systemD systems. I’m not really interested in looking for the reason for these differences, at the level of the memory consumption of individual apps within the OS – I’ll leave that to someone else …
- This reply was modified 1 year, 9 months ago by mikey777.
- This reply was modified 1 year, 9 months ago by mikey777.
- This reply was modified 1 year, 9 months ago by mikey777.
▪ 32-bit antix19.4-core+LXDE installed on :
- (2011) Samsung NP-N145 Plus (JP04UK) – single-core CPU Intel Atom N455@1.66GHz, 2GB RAM, integrated graphics.
▪ 64-bit antix21-base+LXDE installed on:
- (2008) Asus X71Q (7SC002) – dual CPU Intel T3200@2.0GHz, 4GB RAM. Graphics: Intel Mobile 4 Series, integrated graphics
- (2007) Packard Bell Easynote MX37 (ALP-Ajax C3) – dual CPU Intel T2310@1.46GHz, 2GB RAM. Graphics: Silicon Integrated Systems.sybok
Hi, if you are interested in the total memory consumption only, the free utility is probably the best thing.
Redirecting has a nice output, e.g.
free > mem_check_$(date | sed -u 's/ /_/g').txt
provides a file with date in the title.
This can be misleading.
E.g. if some advanced/user-friendly/polished distribution, such as Ubuntu, starts to run check for updates in the background, the memory consumption increases without you ever knowing.
If you measure the memory consumption with/without it running (and not taking this into account), you may end with misleaded conclusions.
That’s why a more detailed examination was suggested in this thread.Also, beware of differences – among several tools – in reporting memory and preferably use a single one.
rokytnji
Systemd trashes /usr/local/bin folder in antix since systemd defaults to using /bin to launch apps.
Simple enough? I have disabled the line in systemd00 text file with hash mark to test with a dist update and dist-upgrade to test what happens when you turn off the preferences.d file inside of /etc/apt/.
1st thing lost is antiX goodies. Control center will show up. None of the buttons will work anymore though.
Sometimes I drive a crooked road to get my mind straight.
Not all who Wander are Lost.
I'm not outa place. I'm from outer space.Linux Registered User # 475019
How to Search for AntiX solutions to your problems
wildstar84
@olsztyn, et. al.: My guess is that the memory footprint differences on “more powerful” hardware is likely due to additional drivers specific to that hardware being loaded. To confirm or disprove this, after booting up both systems but before manually launching other stuff, do:
lsmod | sort >/tmp/lsmod.outand compare between systems.
My guess is that the memory footprint differences on “more powerful” hardware is likely due to additional drivers specific to that hardware being loaded.
@ wildstar84:
Thanks much. I will try to identify the differences. It is not an issue however, considering newer machines usually are expected to have more memory, so additional memory use is not detrimentally affecting performance.
I know it is self-understood, however…
Thanks and Regards…Live antiX Boot Options (Previously posted by Xecure):
https://antixlinuxfan.miraheze.org/wiki/Table_of_antiX_Boot_ParametersAnonymous
yes a 64 bit will use more memory than a 32 bit. please state which version using
and core, base, full. on versus use a comparable core/base/full on other distros.
I:E .. antiX-core isn’t equal to Ubuntu-full. and comparable kernel versions for
system D. Is more of a matter of freedom and choice too.ModdIt
Memory footprint on modern multi core devices is inevitably higher due complex process scheduling and load balancing requirements.
Cores are dynamicaly scheduled to avoid bogging down process or having extremely uneven die temperatures which would occur if a
single core was heavily loaded while others idle.Generaly on these kind of devices we also have plenty of memory, in my case 8GB. Luxury pure for antiX, I bought this machine
long ago but to me it is the most modern I own. Many devices now often have 8 or more cores and 16 to 32 GB Memory, windoze needs
it I am told.Maybe the fottprint is minimaly higher with system D inscessant logging in the background. Never tried to compare, I refuse to use D
because of its all pervading nature, lunatic dependency chains and gigantic logfiles that are not plain text. Will not go back for sure.- This reply was modified 1 year, 8 months ago by ModdIt.
- You must be logged in to reply to this topic.
