Minimum System Requirements

[PK]

Apparently, zRam is still in the experimental (staging) phase for the Linux
kernel, although it seems that Google is going to apply it in Chrome OS.

A much simpler measure for increasing performance on low memory systems, is
a decrease of swappiness (or swap use):
https://sites.google.com/site/easylinuxtipsproject/first-xubuntu#TOC-Decrease-the-swap-use-very-important-

This has a less powerful effect than zRam, but at least it has the benefit
of being straightforward and not complicating things.

As said, my limited experience with zRam is positive. But I know of at
least one person who reported stability issues because of zRam (they
appeared after installing zRam, and disappeared after removing it).

Maybe a swappiness decrease can be considered for Xubuntu instead?

Regards, Pjotr.


2013/7/17 Micah Gersten <micahg at ubuntu.com>

>  On 07/17/2013 01:37 AM, Elfy wrote:
>
> On 16/07/13 21:18, Ali Linx (amjjawad) wrote:
>
>  ...
>
>> I think it's pretty late in the cycle for a major change like this,
>>
>
>  Hmm, I'm sorry to ask too much and cause headache but, what makes you
> think it is pretty late in the cycle
>
> The cycle started months ago.
>
>   and what makes you think this is a major change? why to wait for the
> next cycle?
>
>
> why do it now?
>
>  ...
>
>  I can do this right now. Unless of course everyone else agrees with you
> that it is too late for this cycle and this needs to be discussed later.
>
>
>    I'd not vote on anything that isn't presented to us with all the
> information we need to make informed decisions, in the way that you have
> been asked to do so.
>
>  ...
> Hmm, maybe if I explain more about how this change will help Xubuntu a
> lot, will that change anything about trying out this now? since we are
> still in Alpha Stage. Please, advise!
>
>
>
> I'd certainly be interested to see what we would gain from using it, but
> that should be part of the proposal.
>
>
>
> The reason to do it now is that the next cycle is an LTS cycle which we
> almost certainly wouldn't enable such a thing.  As zram-config just adds an
> upstart job, it's easy enough to pull out of the default install if it
> appears to be problematic.  I'd certainly like to see the proposal (unless
> we can just read the lubuntu thread and start a similar discussion here).
> If it's experimental, then I agree that we shouldn't be enabling it in the
> default install.  If it's just additional, but stable, then it's something
> that can be considered.  We have 1 month until feature freeze, so anything
> that we think might be worthy for the LTS that we can throw in now *and
> stabilize by feature freeze*, we should consider doing so to get extended
> testing on it (read: an extra 6 months).
>
> Micah
>
> --
> xubuntu-devel mailing list
> xubuntu-devel at lists.ubuntu.com
> https://lists.ubuntu.com/mailman/listinfo/xubuntu-devel
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.ubuntu.com/archives/xubuntu-devel/attachments/20130717/1afb7932/attachment.html>