[snip]
>> If I throw out installing a separate OS on a separate machine for each
>> workstation and all of the proprietary thin-client protocols, I think
>> I have 3 options:
>>
>> 1. Connect monitors, USB keyboards, and USB mice directly to a server
>> with multiple video cards.  I found a motherboard with 6 PCI-E slots:
>>
>> http://www.newegg.com/Product/Product.aspx?Item=N82E16813128508
>>
>> 6 video cards could be installed for 6 workstations if the server goes
>> headless, and even more if multi-headed video cards are used.  Xorg
>> requires some special configuration for this but this discussion from
>> 2010 sounds like it's something that is actually done:
>>
>> http://forums.gentoo.org/viewtopic-t-836950-start-0.html
>>
>> These guys got it working in 2006:
>>
>> http://www.linuxgazette.net/124/smith.html
>>
>> 2. Set up a separate thin client for each workstation and run LTSP on
>> the server.  This seems inferior to #1 because it requires setting up
>> and maintaining the LTSP server and client configuration, NFS, xinetd,
>> tftp, dnsmasq, and PXE-boot.  Bandwidth would also be limited compared
>> to #1 and hardware and power requirements would be much greater.
>>
>> 3. Run a Plugable thin client for each workstation:
>>
>> http://www.amazon.com/dp/B004PXPPNA
>>
>> This likely requires running "Userful Multiseat Linux" on my server
>> which is only packaged up for Ubuntu.  The Plugable thin client
>> connects to the server via USB 2.0 which makes me wonder if it could
>> be made to work without Userful Multiseat Linux as a USB video card
>> and input devices, but I imagine drivers for the video card and
>> bandwidth over USB could be a problem.
>>
>> I think #1 is the way to go but I'd love to hear anyone else's opinion
>> on that.  Has anyone here ever set up multiseat in Xorg?
>
> Can you rely on Xorg devs to ensure that they are not going to break your
> multiseat system in the future?

Maybe I'm wrong, but I don't know why there would be (much) more
likelihood of regression with Xorg multiseat than with anything else,
including LTSP and all of its dependencies.  In the context of both
hardware and software, I think there are much fewer points of
potential failure with multiseat than with an LTSP thin-client
arrangement.

> Are you sure that you will come across bandwidth issues if you follow option
> #2?  On a gigabit network at work we're running thousands of thin clients
> distributed across hundreds of VM servers, and there is no noticeable latency
> (unless a particular VM MSWindows server plays up).

I'm sure I wouldn't.  I only mentioned the increased bandwidth of
multiseat vs. thin-clients as a technicality.

> I understand that managing multiple boxen is always a greater burden, but
> something like GNAP may lighten the work needed?
>
>  http://www.gentoo.org/proj/en/base/embedded/gnap-userguide.xml

That looks cool, but from my perspective it's another layer to learn,
install, configure, and manage.  chef and puppet take a different
approach to lessening the burden of administrating multiple systems,
but in the end neither approach comes anywhere near the hardware and
software simplicity (and corresponding ease of setup and maintenance)
of multiseat.

- Grant

Reply via email to