[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: [K12OSN] Re: MOSIX



Excellent!

Error checking would be nice on the scripts but if it's installed on a clean
LTSP 2.0.x system it shouldn't be an issue (I wouldn't imagine.  Sounds like
you've been busy! :-)  Looking forward to your package(s).  Be sure to
submit your documentation to Moshe Bar so he can post it on the OpenMosix
site.

Ah, Now I see you.  You are putting your cluster to work with that -j
option.  Nice.  Do you have any scripting for controlling which process
migrate for particular apps?  I think you mentioned something about this a
few weeks back.

(Gee, I hope people aren't getting tired of all the OpenMosix stuff, but I
cannot think of a better project for a school running K12LTSP to test out
then this one!!)

You wrote:
> Room for the nodes is an issue.  Running out of power is a bigger one

>   On the positive side you'll have a nicely heated room in the winter.

Yeah, that's my big joke with a friend from work, once I get the cluster set
up I will have to buy a bigger air conditioning unit.  Winters should be
nice though.  No.  Seriously, I cannot run 20 systems at home.  I don't have
the room or power for it.  I will have to find some space elsewhere for the
demo/testing.  Perhaps the local LUG can help.  Or maybe, better yet, a
school!

> Adding the information from the howto is fine.

Thanks Richard.  I'll try to get it put together here in the next few days.
I think it will make the step-by-step flow more smoothly, but I still think
your original on MOSIX should be a must read, or better yet, IMHO, a new
detailed How-To on OpenMosix that explains what each step is doing.

James


----- Original Message -----
From: "Richard Camp" <rcamp campworld net>
To: "James Jensen" <jmsjnsn linuxfreemail com>
Cc: <K12OSN redhat com>
Sent: Thursday, April 04, 2002 11:51 PM
Subject: Re: [K12OSN] Re: MOSIX


> James,
>
> My test cluster is now running OpenMosix.  Install scripts are going
> well.  The scripts are brute force right now(aka no error checking).
> I'm currently working on a pre compiled ltsp openmosix kernel.  My first
> package will be a .tgz file and/or iso image.  Everything that is needed
> from openmosix and ltsp will be included.  The package size looks like
> it will be 45-50meg including kernel sources.
>
> My test cluster is as follows:
> server
> PII400 (dual cpu)
> 256M RAM
> hd, cdrom
> 2 100mbit cards
>
> clients (4 running currently)
> 366 celeron
> 64-128Meg ram
> 100mbit net cards
>
> Switch
> 8 port
> 100mbit full duplex
>
> The -j option is the number of jobs make should run.  The files that
> take longer to compile get migrated.
>
> Room for the nodes is an issue.  Running out of power is a bigger one

>   On the positive side you'll have a nicely heated room in the winter.
>
> Adding the information from the howto is fine.
>
> Richard
>
> James Jensen wrote:
>
> > Richard:
> >
> > Now that's good stuff!  I'm looking forward to your continued
documentation
> > of this stuff.  Were you running 100Mbit FD or HD?  What would you
estimate
> > as "smaller" installation?  I am hoping some of the labs out there get
to
> > test this stuff and post some stats.
> >
> > I intend on building a AMD XP Dual-Processor box with 1-2GB RAM for
testing
> > and demonstration purposes, and I hope to collect enough "junk" systems
for
> > about 20 clients.  Where I will put this stuff I don't know yet but
we'll
> > see what we can do.  I will have to go with Fast Ethernet, I won't be
able
> > to afford Gigabit at this point.
> >
> > What does the -j6 switch do for us during the kernel compile?
> >
> > I know you are putting together documentation, scripts, etc as you can.
I
> > wonder if you might grant permission for me to update my Mini How-To
with
> > the cp {userland tools} section of your Mosix text?  At this point I
feel
> > like it's kind of lacking a little without that completion.  Do you
mind?
> >
> > Thanks,
> >
> > James
> >
> >
> > ----- Original Message -----
> > From: "Richard Camp" <rcamp campworld net>
> > To: "James Jensen" <jmsjnsn linuxfreemail com>; "Guy Lessard"
> > <glessard coll-outao qc ca>; <K12OSN redhat com>
> > Sent: Thursday, April 04, 2002 2:13 AM
> > Subject: Re: [K12OSN] Re: MOSIX
> >
> >
> >
> >>Everyone,
> >>
> >>>From my testing 8 nodes running seti home went pretty good.  100Mbit
> >>
> > switch should be fine for smaller installations.  A switch with a
> >
> >>gigabit link to the server should keep most people happy for some time.
> >>
> > The idea of course is to make the client display responsive.
> >
> >>You may want to try traffic shaping.  Have the server kernel give the
> >>
> > highest traffic priority to the Xwindows protocols.  Your client display
> >
> >>refresh rate will rely on how loaded the server is.
> >>
> >>The next challenge is migrating processes off the server.  The easiest
way
> >>
> > I've found is to tell mosix that the CPU speed is slower than it
> >
> >>really is.  This has worked for me at least.
> >>
> >>The other issue is process execution time.  The longer the process runs
> >>
> > the more likely it will migrate to another node.  The next time you
> >
> >>compile the kernel do the following.
> >>
> >>Take the number of nodes you have and add one to it.  If you have any
> >>
> > nodes that have multiple CPUs, count each CPU as a node.  My
> >
> >>example cluster gives me 6.  Type
> >>
> >>make -j6 bzImage
> >>make -j6 modules
> >>
> >>Watch the load on the cluster.  Your mileage will vary on this one:)
> >>
> >>Good luck
> >>Richard Camp
> >>
> >>
> >>>I know one thing for sure, with the 4-5 PC's I had running on 10Mbit,
> >>>
> > that
> >
> >>>poor little hub was smokin' when I would stress test the cluster--solid
> >>>green activity lights across the board.  The collision light wasn't
> >>>
> > solid
> >
> >>>but it was flashing *a lot*.  Watching the lights is not a very
accurate
> >>>measure of traffic of course, LOL, but with no management facility on
> >>>
> > it,
> >
> >>>that's the best I tell you.
> >>>
> >>>When the cluster was under load it still appeared that the clients
> >>>
> > Xwindows
> >
> >>>refreshes were working well (even better then  with non-OpenMosix
> >>>
> > kernels.
> >
> >>>Especially on that old 486-66mhz.  It really made it feasible &
> >>>
> > worthwhile
> >
> >>>to use that old dog as a client.).
> >>>
> >>>As you mentioned it will be very interesting to see what happens on a
> >>>
> > large
> >
> >>>set up.  I'm guessing that a Fast Ethernet (100Mbit Full Duplex)
network
> >>>would be able to handle the load even on a larger install.
> >>>
> >>>The big concern, to me, remains--what happens when Client #1 process A
> >>>
> > is
> >
> >>>running on Client #5, and Client #5 powers off.  What happens to
process
> >>>
> > A?
> >
> >>>The failover factor...
> >>>
> >>>It may just be that OpenMosix would need to be a cluster of servers
only
> >>>
> > at
> >
> >>>this point.
> >>>
> >>>Valid questions all, and it would be great fun to test this stuff in a
> >>>
> > real
> >
> >>>lab.
> >>>
> >>>James
> >>>
> >>>----- Original Message -----
> >>>From: "Guy Lessard" <glessard coll-outao qc ca>
> >>>To: "James Jensen" <jmsjnsn linuxfreemail com>
> >>>Cc: <K12OSN redhat com>
> >>>Sent: Thursday, April 04, 2002 8:23 AM
> >>>Subject: Re: [K12OSN] Re: MOSIX
> >>>
> >>>
> >>>
> >>>>Boy all this mosix-k12ltsp stuff is very interesting.
> >>>>I was wondering what will be the impact of having processus eating up
> >>>>
> > the
> >
> >>>>bandwith of the network, when will the screen-keyboard refresh start
> >>>>
> >>>hurting
> >>>
> >>>>because of all these processus flying around?
> >>>>I guess this is a early question but a large installation of this
> >>>>
> >>>mosix-k12ltsp
> >>>
> >>>>setup would start giving us answers.
> >>>>
> >>>>






[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]