[Sugar-devel] [IAEP] Getting data about the upgrading older machines and SoaS responsiveness.

[Tomeu Vizoso]

On Mon, Jun 8, 2009 at 13:03, James Zaki<james.zaki at gmail.com> wrote:
> With regards to the speed issue.
>
> I tried SoaS on a USB2.0 (but not high-speed) memory-stick, performance was
> hideous on a macbook.
> Using a USB2.0 high-speed memory-stick, performance is great on an eeepc,
> which has 1G Ram. I know its not small, but its all I have to compare with
> for now.
>
> So from what I have experienced the USB port would be the first target. I'll
> hopefully get a chance to test on low-RAM school computers tomorrow.

We should see why disk I/O is affecting so much performance. Instead
of making the disk faster we should make Sugar stop doing stupid
things (if that's indeed the case).

Anybody with performance profiling of linux system can give some hints
about which tools we can use to see why disk I/O seems to be the
bottleneck?

Thanks,

Tomeu

> James.
>
>
>
> 2009/6/8 Jonas Smedegaard <dr at jones.dk>
>>
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: RIPEMD160
>>
>> On Sun, Jun 07, 2009 at 07:00:28PM -0400, Luke Faraone wrote:
>> >On Sun, Jun 7, 2009 at 18:43, Jonas Smedegaard <dr at jones.dk> wrote:
>> >
>> >> >It sound like another great, low impact (I am trying to think of a
>> >> >term like 'carbon foot print' to properly reflect the impact) way of
>> >> >bringing LTSP into the class room.
>> >>
>> >> "polite" or "gentle" perhaps?
>> >>
>> >> or "non-invasive"?  Emphasizing what is avoided: invading -
>> >> potentially taking over, accidentally or on purpose, the computers
>> >
>> >
>> >Granted, you would *need* to check with your local systems
>> >administrator before implementing LTSP. (as opposed to a lower-risk
>> >USB-local-booting solution) At my school, for example, netbooting a
>> >workstation starts the "recloning" process of loading a new Windows XP
>> >image; setting up LTSP without asking would cause major problems with
>> >their work.
>>
>> non-invasive to the _computers_ but invasive to the network
>> infrastructure.
>>
>>
>> So yes, a better term would be good, to not risk sysadmins feeling
>> cheated when learning the hard way that this so-called non-invasive
>> system includes a DHCP daemon, breaking their WiFi hotspots, printers
>> and what not.
>>
>>
>>  - Jonas
>>
>> - --
>> * Jonas Smedegaard - idealist og Internet-arkitekt
>> * Tlf.: +45 40843136  Website: http://dr.jones.dk/
>>
>>  [x] quote me freely  [ ] ask before reusing  [ ] keep private
>> -----BEGIN PGP SIGNATURE-----
>> Version: GnuPG v1.4.9 (GNU/Linux)
>>
>> iEYEAREDAAYFAkosXD0ACgkQn7DbMsAkQLgTYwCeNg687lF4eEXrGw9SqB62AGih
>> 5WQAniL/ZEmBKsZ8zVMCRmPlNnScHmE5
>> =lu5m
>> -----END PGP SIGNATURE-----
>> _______________________________________________
>> Sugar-devel mailing list
>> Sugar-devel at lists.sugarlabs.org
>> http://lists.sugarlabs.org/listinfo/sugar-devel
>
>
> _______________________________________________
> Sugar-devel mailing list
> Sugar-devel at lists.sugarlabs.org
> http://lists.sugarlabs.org/listinfo/sugar-devel
>
>