Seems the DB is once again slow, per the news.....

"September 22, 2004
Both the transitioner and the validator have fallen behind. We are determining whether this situation is because of a software or hardware bottleneck. In the meantime we need to allow these programs to catch up so the scheduler is off over night. We are in the process of obtaining faster hardware for DB server. "

The thing I've got to ask about is the last part: "We are in the process of obtaining faster hardware for DB server."

Didn't we just hear that when they were waiting for the SNAP box? I guess this means the SNAP box failed so badly that they are looking for another solution....

They could have old computers in a storage room somewhere. As far has I know they don't show or say very much or anything about their gear.I think sometime ago they said that they did not know exactly what was needed for the project and would get some better equipment later. Hope they do, because that's quite a complex project to manage .

>
> They could have old computers in a storage room somewhere. As far has I know
> they don't show or say very much or anything about their gear.

You must be a newbie, as the SNAP box was brought in (new) to solve the DB IO bottleneck a month or so back (SNAP 18000 IIRC). They had SNAP folks on site right before labor day, as things kept failing..

From the site news:

September 3, 2004
The SnapApp folks were here for most of the afternoon. They replaced the processors with ones that are faster and have larger caches. More importantly, we talked about how to stabilize the system. The plan is to simplify both the HW and service mix as much as possible in the hopes of removing the variable that is causing the hangs (and of course they will be going for the root cause at their lab). We altered the startup so that a bare minimal set of services are running. And we are going to only use the 18000 head unit for now, keeping no data on the sd30 expansion. This means moving some data around. We first have to move the upload/download data to a temporary volume on the sd30, then reconfigure the space on the 18000 head, then move the uldl to the head. The move to the temp volume is happening now. Tomorrow morning we will do the reconfiguration and start final move. With luck we will have the project up by EOD tomorrow.



July 23, 2004
The good news is that the scheduler is now running as a fast cgi which means it maintains a persistent connection to the DB. The load on the scheduling server has gone way down which is a good thing and the mysql server is using more CPU which is also a good thing. The bad news is that the transitioner is still way behind and so the project will probably run out of send-able work as early as 04:00 UT tomorrow. The transitioner is slow because the mysql server is IO bound. We are having Sun look at our IO system to see if it can be tuned to better performance. At the same time we are planning on moving the physical DB to a high performance Snap Appliance box in the near future. So we are hopeful that these IO problems can be solved.

(look at the last sentence)

> >
> > They could have old computers in a storage room somewhere. As far has I
> know
> > they don't show or say very much or anything about their gear.
>
> You must be a newbie, as the SNAP box was brought in (new) to solve the DB IO
> bottleneck a month or so back (SNAP 18000 IIRC). They had SNAP folks on site
> right before labor day, as things kept failing..
>
> From the site news:
>
> September 3, 2004
> The SnapApp folks were here for most of the afternoon. They replaced the
> processors with ones that are faster and have larger caches. More importantly,
> we talked about how to stabilize the system. The plan is to simplify both the
> HW and service mix as much as possible in the hopes of removing the variable
> that is causing the hangs (and of course they will be going for the root cause
> at their lab). We altered the startup so that a bare minimal set of services
> are running. And we are going to only use the 18000 head unit for now, keeping
> no data on the sd30 expansion. This means moving some data around. We first
> have to move the upload/download data to a temporary volume on the sd30, then
> reconfigure the space on the 18000 head, then move the uldl to the head. The
> move to the temp volume is happening now. Tomorrow morning we will do the
> reconfiguration and start final move. With luck we will have the project up by
> EOD tomorrow.
>
>
>
> July 23, 2004
> The good news is that the scheduler is now running as a fast cgi which means
> it maintains a persistent connection to the DB. The load on the scheduling
> server has gone way down which is a good thing and the mysql server is using
> more CPU which is also a good thing. The bad news is that the transitioner is
> still way behind and so the project will probably run out of send-able work as
> early as 04:00 UT tomorrow. The transitioner is slow because the mysql server
> is IO bound. We are having Sun look at our IO system to see if it can be tuned
> to better performance. At the same time we are planning on moving the physical
> DB to a high performance Snap Appliance box in the near future. So we are
> hopeful that these IO problems can be solved.
>
> (look at the last sentence)
>


Bird says be nice Woodrow........
My name is Pascal and this message has my approval...

It is 10 oclock, do you know what your WUWUs are doing tonight...

> Bird says be nice Woodrow........
> My name is Pascal and this message has my approval...

Bird says Pascal is lost and doesn't see a valid question unless it's written in bird poop on Pascal's head!

My name is azwoody, and I approve the fact that Pascal is lost!

>
> > Bird says be nice Woodrow........
> > My name is Pascal and this message has my approval...
>
> Bird says Pascal is lost and doesn't see a valid question unless it's written
> in bird poop on Pascal's head!
>
> My name is azwoody, and I approve the fact that Pascal is lost!
>
>
I must say LOL :D
In that i can agree, Pascal is like a Volture (abc check) who attacks every kind of thread and chews it rotten.
I'm sorry, but have to let it out.

> I must say LOL :D
> In that i can agree, Pascal is like a Volture (abc check) who attacks every
> kind of thread and chews it rotten.
> I'm sorry, but have to let it out.


Thank you!

>
> > I must say LOL :D
> > In that i can agree, Pascal is like a Volture (abc check) who attacks
> every
> > kind of thread and chews it rotten.
> > I'm sorry, but have to let it out.
>
>
> Thank you!
>
> No problem

>
> > Bird says be nice Woodrow........
> > My name is Pascal and this message has my approval...
>
> Bird says Pascal is lost and doesn't see a valid question unless it's written
> in bird poop on Pascal's head!
>
> My name is azwoody, and I approve the fact that Pascal is lost!
>
>


Attacks?????, now now, Bird says be nice :o)We know where you are going with this Woodrow..........
My name is Pascal and this message has my approval...

It is 10 oclock, do you know what your WUWUs are doing tonight...

> >
> > > I must say LOL :D
> > > In that i can agree, Pascal is like a Volture (abc check) who
> attacks
> > every
> > > kind of thread and chews it rotten.
> > > I'm sorry, but have to let it out.
> >
> >
> > Thank you!
> >
> > No problem
>
>


Learn to spell wiz ace........;o)or get a dictionary, like Webster's New World Compact Office Dictionary, I have one........
My name is Pascal and this message has my approval...

It is 10 oclock, do you know what your WUWUs are doing tonight...

>
> Learn to spell wiz ace........;o)
> My name is Pascal and this message has my approval...
>
> It is 10 oclock, do you know what your WUWUs are doing tonight...
>

What are you trying to add to this thread Pascal, other than to prove that you can be a royal pain in the butt?

Are you adding info, or are you just attacking someone you don't agree with?

Seems "Troll" comes into play here.......

> > >
> > > > I must say LOL :D
> > > > In that i can agree, Pascal is like a Volture (abc check) who
> > attacks
> > > every
> > > > kind of thread and chews it rotten.
> > > > I'm sorry, but have to let it out.
> > >
> > >
> > > Thank you!
> > >
> > > No problem
> >
> >
>
>
> Learn to spell wiz ace........;o)or get a dictionary, like Webster's New World
> Compact Office Dictionary, I have one........
> My name is Pascal and this message has my approval...
>
> It is 10 oclock, do you know what your WUWUs are doing tonight...
>
>
>ADDED (abc check) RECALL?!?

> They could have old computers in a storage room
> somewhere. As far has I know they don't show or
> say very much or anything about their gear.

I found these photographs of their equipment.
http://totl.net/Spud/gallery.html



Rocky

<a> [/url]

Ironic isn't it - the server that seems to carry many of the BOINC processes is named "klaatu".

Now for those that don't recall, 'Klaatu' was the 10 foot metal monster from outer-space in the "The Day Earth Stood Still" - Seems more a case of "The Day the Server Stood Still" :-)

Useless fact No. 37

> Ironic isn't it - the server that seems to carry many of the BOINC processes
> is named "klaatu".
>
> Now for those that don't recall, 'Klaatu' was the 10 foot metal monster from
> outer-space in the "The Day Earth Stood Still" - Seems more a case of "The Day
> the Server Stood Still" :-)
>

You're very close! :D Klaatu was the alien that first came out of the flying saucer. Gort was the robot.

Ah yes, the unforgetable line... "Gort, Klaatu barada nikto"

I don't think we ever got a literal interpretation of that, but it was the only thing that stopped Gort from destroying the Earth. Man, what a flick... gotta watch it again sometime soon! :)

Edit: Another great line from Klaatu was "I'm impatient with stupidity. My people have learned to live without it". Considering some of the posts we've had lately around here... it seems even more appropriate for a validator name. ;)

Oh so close.... Googled to confirm what I thought I already knew :D

Should have read one or two of the results - not just the summaries :-(

And, yes, must keep an eye-out for a re-run of the film - a good one.