PDA

View Full Version : Getting work to do



Chuck
04-20-2008, 12:17 PM
All,
With the recent tweaks to the windows OS by Alpha, there clearly isn't enough work being metered out of the Server.


If memory serves, someone here knows how to get in contact with Lijun.

If so, and we agree, can we please ask Lijun to set the server to allow more work to be distributed (jobs/hr or jobs/user or jobs/day)? If the server is at its max, then perhaps they can slave on another machine or something?

It seems that Eon, as it should, doesn't take a lot of CPU resources to run and if someone wanted to push extra resources... then 2 or 3 cores of 3-6 instances should be enough, but even here, with my DSL, it runs great for a while, then starves waiting for work.... This is a far cry better than being unable to return work (as it was in the past).

Can this be brought to Lijun's attention promptly? I think we as FDC can gain a lot and Eon can get a lot more work done.

Thoughts???

Thanks,
Chuck

Paratima
04-20-2008, 09:33 PM
Sent. Will post reply. :thumbs:

Having said that, let me point out that answers to this type of question have been posted before. Essentially, (as well as I understand it) the parameters of the current situation are flung out to the troops. Answers come back and are applied to the problem. Fine, so far. Rinse and repeat indefinitely. Then the server gets to a point where it has to figure out a way "over the hump". Processing slows. If it's a major hump, processing stops. ZZZ's happen. Impatience ensues. Sometimes, the server gets SO hung up that only a kick in the butt will get it going again. Most times, it gets there by itself, although it does seem to slow down overall with time and eventually requires a reboot to get the ball rolling.

The situation is vastly improved over the pre-Lijun period. He has really improved the situation, but it appears to be limited in scope, possibly by the way the problem is being approached. In other words, it might be possible to get a quantum leap, but at this stage it might require a major rewrite.

I have forwarded your concerns and also asked for a project progress update, as we haven't heard anything in quite a while.

Stand by. :D

Chuck
04-20-2008, 09:54 PM
Paratima, thanks for sending out the email promptly.... Greatly appreciated! I guess the decision to inquire to Lijun was a nobrainer. :) :cheers:

I noticed a lot more work starvation today, but in total, it seems as if we all are getting more work done. Bravo, and thank you Alpha!!! :banana:

This is one of those times when I wish I knew/remembered the win32 internals.

Alpha, again thanks for being there with the knowledge and right tools for the job at the right time as well as the 'manual way' (love working with regedit) of getting it done. I use PC-Tools other products, but not this one yet... guess now I will be needing it as I lose that knowledge/memory of the internals.

Thanks to EVERYONE who has been keeping their eyes open and looking in documentation to find the parameter names and values. I suspect this will have impact on MANY of our projects!!!!!

Good night all.... I look forward to some great news about Eon improvements as well as our other DC projects. I smell some great impact on other projects and can't wait to 'catch the other teams with their pants down'... :)


Cheers,
Chuck

Paratima
04-20-2008, 10:01 PM
Interestingly, I checked my registry and found that the change to 30 seconds had already been made!

Then I remembered that, using Firefox as my browser, I had installed a helper app named, ingeniously, Tweak Network Settings. Ran it once after installation, selected "Max Everything". (Well, of course!) And never thought about it again. Seems to have made that adjustment, among others.

Will be interested at Lijun's response - also cc'd Graeme (Dr. Henkelman) who is the overall project admin and who wrote the original code, btw. He it was who brought EON from its birthplace in UWashington to UTexas.

Paratima
04-20-2008, 10:08 PM
PS. Ironbits and I pretty much agree on this about EON:

1. Put as much horsepower on it as you feel moved to do.
2. Make sure everything's running nicely.
3. Turn your monitor's ON/OFF switch to the OFF position
4. Go play with your kids/dog/cat/wife/motorcycle/whatever. DON'T watch it and you'll be happier.
:rotfl:

alpha
04-21-2008, 03:34 AM
Excellent, I'm glad to hear it if you're experiencing improvements since making that change to the registry. I'm not sure if a reboot is required or not after the tweak. I don't think I've rebooted yet so I can't really comment on whether I'm seeing any changes or not. Actually, I'll reboot now.

Of course, it would be great if we could get less ZZZs and jump on this as a primary project instead of a side runner. We could then maybe push for top place. I use a tool called SetPriority (http://www.free-dc.org/forum/showthread.php?t=6388&highlight=setpriority) by Digital Parasite to make sure eOn gets the cycles when work is available.

Edit: the reboot seems to have done the trick, I'm seeing the sockets closing much quicker now.

Paratima
04-21-2008, 07:55 PM
OK, here's the best I can do at the moment.
First, see this link (http://www.free-dc.org/forum/showthread.php?t=4928), which describes the start-up.

Then see 2nd link (http://www.free-dc.org/forum/showthread.php?t=12274), which talks about the changes made a year ago.

See my feeble description of the processing cycle above.

After I passed along your query, I got this:
Hi Les,
We are getting a new graduate student to work on this EON project. It is gonna take some time for him to get used to it. Thanks for advice.
Lijun

So, some days, you're the pigeon; some days, you're the statue. Progress is sometimes just about as slow as the processes being calculated. Could be wrong, but I read the response as being, "We're reasonably happy with what we've got and any changes are going to take a while."

They may not have the resources to do any better right now. As I explained above, it's not particularly a lack of processing power as much as the cycle of the processing algorithm. Not knowing any more about the overall scheme of things makes it hard to make meaningful suggestions. I, and others, have found some correctable items in the past, but unless you're there...

movieman
04-22-2008, 01:00 PM
If you want to make EON "crank" do a ton of installs.
One day for chuckles I decided to see if I could match that 'Welltest" account which is their computer lab with 75 dual cores.
Tossed 2 clovers, a Q6600 and an old FX51 at it.
60 installs on each of the clovers, 35 on the Q6600 and 7 on the FX51 to max out the machines..
Yup, I got 'em that day..:rotfl:

IronBits
04-22-2008, 03:43 PM
I had over 150 clients running there for a while about a year ago. :D

em99010pepe
04-22-2008, 04:02 PM
I had over 150 clients running there for a while about a year ago. :D

And today? Try to match Mercenaries output...

LAURENU2
04-22-2008, 04:03 PM
I have 90 clients running right now one on each node here
And weltest is still doing 7 X my output
And even when I had 300 clients running here 3 per node
I only came close to 80% of his out put
movieman Can you do a repeat of this match for us:Pokes:

em99010pepe
04-22-2008, 04:04 PM
I have 90 clients running right now one on each node here
And weltest is still doing 7 X my output
And even when I had 300 clients running here 3 per node
I only came close to 80% of his out put
movieman Can you do a repeat of this match for us:Pokes:

Install 13 clients per core...This is what I have:

1 x AMD 64 with 13 instances
1 x P4 with 13 instances
1 x Q6600 with 52 instances

LAURENU2
04-22-2008, 04:24 PM
Install 13 clients per core...This is what I have:

1 x AMD 64 with 13 instances
1 x P4 with 13 instances
1 x Q6600 with 52 instances
Uesing your numbers
I need to start up 2,041 clients here
I do not think EoN has the power to keep up with a push like that

em99010pepe
04-22-2008, 04:27 PM
Uesing your numbers
I need to start up 2,041 clients here
I do not think EoN has the power to keep up with a push like that

I think the server can handle the increase of connections....Start having 6 clients per core and wait to see what happens.

LAURENU2
04-22-2008, 05:55 PM
I think the server can handle the increase of connections....Start having 6 clients per core and wait to see what happens.

I am getting ZZZZZ that make each My PC's Beg for 3 to 5 min 's for new work:swear:
If I start up 900+ clients((6 Per core)) that wait time will jump to 5 to 8 Min's for all members:clap:

From what I have seen EoN does not have the Power to supply the needs of it's membership as it stands right now :cry:
Adding the power you are suggesting would only I think would Overwhelm the server to the point that no one will be happy with the Project :swear:

When I first jumped in EoN at full speed at only 3 per core that is just what happened the wait time went to 10 min per client

Waiting like that to me is a Waste of CPU time and power Not to mention Bandwidth

I think until the Admins :Pokes: upgrade or streamline there there Server load
I will keep it at 1 per node not core:hifi:

If the EoN's :Pokes:Admins :Pokes:are reading this I hope you can see my points and make improvements to there system

em99010pepe
04-22-2008, 07:49 PM
You talk too much, bring them on....

Paratima
04-22-2008, 08:00 PM
I think Lauren's right. EON is a self-eating watermelon.

If the boys at UT thought they could needed to get more science done than is happening now, they would find a way to streamline the process. An alternative explanation may be that they are, after all, scientists, NOT programmers. It may be beyond their means to even visualize, much less produce a more effective system.

(No, I don't think very highly of scientists, who are not trained as programmers, writing their own code. Look at Proteins @ home for an example.)

If they want my GHz, they know where to find me. :D

LAURENU2
04-22-2008, 08:47 PM
You talk too much, bring them on....

Thats Funny My Wife Said The Same Thing To Me Last night When I Came to Bed:blush:

LAURENU2
04-26-2008, 04:31 AM
Well the work seems to be flowing Get the WU;s while you CAN:Pokes:

LAURENU2
04-26-2008, 10:51 PM
You talk too much, bring them on....
So now how am I doing :Pokes:
I am at 1 client per core now instead of 1 client per Node I was
I am doing 50% of the teams output Is this what you meant by "bring them on":hifi:

:idea:em99010pepe do you and your DC Mercenaries want to Race to a mill:Pokes:
I Only Have 3 more week of runtime before I have to shut down for a month
Might as well have fun :harhar:with it:harhar:

em99010pepe
04-27-2008, 05:11 AM
So now how am I doing :Pokes:
I am at 1 client per core now instead of 1 client per Node I was
I am doing 50% of the teams output Is this what you meant by "bring them on":hifi:

:idea:em99010pepe do you and your DC Mercenaries want to Race to a mill:Pokes:
I Only Have 3 more week of runtime before I have to shut down for a month
Might as well have fun :harhar:with it:harhar:

You are doing great!
I reserved another thing for my machines and I need at least two weeks to complete it. Then we can race.

Carlos

LAURENU2
04-27-2008, 10:30 AM
You are doing great!
I reserved another thing for my machines and I need at least two weeks to complete it. Then we can race.

Carlos
:banana:Well I :hiya:Guess I WIN then:lmao:

em99010pepe
04-27-2008, 11:07 AM
Don't be sure about that.

LAURENU2
04-27-2008, 11:35 AM
You talk too much, bring them on....
Don't be sure about that.

:rotfl:Now You are the one talking too much, bring them on.... :fight:

em99010pepe
04-28-2008, 11:22 AM
:rotfl:Now You are the one talking too much, bring them on.... :fight:

I have to shutdown (still live with my parents) for a few weeks all my home machines due to high electricity bill.... I might move them to work, still thinking. My problem is that my work contract ends this June and I won't renew it because there's no projects to finance our salaries. Because I want to finish my master degree I'll have access to the lab where I work, and this is my option, and not to move to another unit, in the institution I work, where I have a place to work (two jobs offer). I'll make a pause of 2-3 months to end my master thesis...I think this is a good choice. I can't work and make the thesis at the same time. For the moment I am dedicated to it and I prefer to stay in that way.

Carlos

LAURENU2
04-28-2008, 04:25 PM
OK :thumbs:

LAURENU2
04-30-2008, 05:11 PM
So what is up with this site Any word from anyone
Will they get there site and stats back up ?????

Paratima
04-30-2008, 07:09 PM
Didn't even realize it was down until I read your post. Any idea (anyone) on when it went down?

I've sent an email asking for details - will share any answers I get posthaste.

LAURENU2
04-30-2008, 08:27 PM
Didn't even realize it was down until I read your post. Any idea (anyone) on when it went down?

I've sent an email asking for details - will share any answers I get posthaste.

I think it has off line for 3 or 4 days now last Data Updated : 2008-04-27 03:01:27

It has beem sooooo long I might be in 1st place already or not hehe

Paratima
04-30-2008, 09:29 PM
The EON work server is back up, as you probably know. According to Lijun, "a large job clogged the memory. I cleaned it up and the server started again." :scratch:

I wrote back that part of the problem is we can't get through to the web server, consequently we can't run 3rd-party stats. Hopefully, he'll see that & give it a reboot or whatever.

LAURENU2
06-13-2008, 06:03 PM
I must be Sucking the Work Server Dry :lmao:
Even weltest can't get it up to 30 K today :thumbs:

LAURENU2
07-26-2008, 03:47 PM
I must be Sucking the Work Server Dry :lmao:
Even weltest can't get it up to 30 K today :thumbs:
Do you think the other teams like Team Norway & DPC will give up if I keep up the hogging of WU's from the Server ?
Seems like they were stating a little Push when one of My DSL lines went down

IronBits
07-26-2008, 04:58 PM
Keep on Keeping on :D

Paratima
07-26-2008, 07:43 PM
I'll bug EON Central again. Seems to get a bit of work flowing, but never lasts very long. I'll try anyhow.

LAURENU2
07-26-2008, 10:35 PM
I'll bug EON Central again. Seems to get a bit of work flowing, but never lasts very long. I'll try anyhow.
I sent them Email a week ago but no reply:bang:

Paratima
07-28-2008, 12:13 PM
Yup. I think we're just running up against the natural boundaries of the interactive way EON works. Throwing a ton more CPU power at it, from our side, won't make a squeek of difference. We're limited by how fast the server can do its bit.

I've requested that we at least get an update on where the project stands in terms of its own goals. We haven't had an update on that score in over a year. You like to think that things are being accomplished by your work, that science is being served, but unless they communicate, we just don't know. Frustrating. :bang:

LAURENU2
07-30-2008, 08:43 PM
I guess they ran out of WU's for today.:bang:

Xaverius
07-31-2008, 09:52 AM
Do you think the other teams like Team Norway & DPC will give up if I keep up the hogging of WU's from the Server ?
Seems like they were stating a little Push when one of My DSL lines went down
Well, it's a shame we can see the difference in getting WU's when it happens.

At this moment not many DPC'ers are into eOn because of the lack of getting WU's, maybe Free-DC likes it this way. :cow:
But I agree that some news (and lots of WU's) from the guys of eOn would be appreciated.

alpha
07-31-2008, 10:05 AM
Seconded! News and work units would be very nice. Maybe if enough of us poke the admins we'll get an answer. :Pokes:

Xaverius
07-31-2008, 10:15 AM
Seconded! News and work units would be very nice. Maybe if enough of us poke the admins we'll get an answer. :Pokes:
Maybe we can set up a DC-project for that! :looney:

jasong
08-01-2008, 08:51 PM
The thing that really bugs me is this looks like really useful science. If there is so much processing power that it can't be utilized properly, than they need to find a way to add more servers to do more projects along the same lines. Maybe have a central server that serves as a go-between for the servers crunching the data and the clients requesting it.

LAURENU2
08-30-2008, 05:28 PM
8,719,379---:bigtrain::bigtrain::bigtrain: +++66,383,348
Long Uphill Track

LAURENU2
09-07-2008, 02:29 PM
I think I am running a DOS attack :whistle: I seem to be the only one getting any work:o
I wounder what would happen if I went to full power

Paratima
09-07-2008, 06:09 PM
All I have on it now, 'cause of OGR, is my old POS internal web router. One day it gets almost 500, next day 20. Go figure.

I thought that after OGR, I'd try a few days' running & see what it looks like. But yeah, what you've got looks a LOT like DOS, at least for the rest of us mortals. :eek:

LAURENU2
09-07-2008, 07:42 PM
To tell you the truth The 1 day I pulled off here to work OGR I made less points from OGR
I went back here to attack Welltest here( I was piss that he sucked up 80K) and made 30 K more at OGR Go Figure

Do you think If I keep up this Pressure and add 25% more the Admins will see that they Need to UPGRADE there IN /OUT Server to handle the volume?

Well we will see I will keep adding power to here and OGR Let's see how fast the server can Run

IronBits
09-07-2008, 09:41 PM
Sounds good. Keep up the pressure :cheers:

LAURENU2
09-08-2008, 01:53 AM
Sounds good. Keep up the pressure :cheers:
Adding more pressure :rock: I took Weltest down from doing 80K a day yesterday to 1K a day today:lmao:
I will try to bring more Nodes back home tomarrow and get them back online

LAURENU2
09-08-2008, 02:06 AM
To tell you the truth The 1 day I pulled off here to work OGR I made less points from OGR
I went back here and made 30 K more at OGR Go Figure

Now I am close to Full power here and I Upped my totals another 30K at OGR:confused:
:umm:Why would working Eon add 60K more points in OGR? Weird But I Like it:banana:

Paratima
09-08-2008, 07:54 AM
:umm:Why would working Eon add 60K more points in OGR? Weird But I Like it:banana:
Has to do with the interconnectedness of all things. :beep:

LAURENU2
09-08-2008, 08:57 AM
Has to do with the interconnectedness of all things. :beep:
OH I must have the new ZEN computers that are one with the world :lmao::rotfl:

[NGS]Cpt00Kirk
09-08-2008, 10:18 AM
:bang::bang: LaurenU2 ... stop taking all the Wu's :bang::bang:

give us a change to outproduce you!! :thumbs:

LAURENU2
09-08-2008, 07:24 PM
Cpt00Kirk;126521']:bang::bang: LaurenU2 ... stop taking all the Wu's :bang::bang:

give us a change to outproduce you!! :thumbs:

I'm not even up to full power yet I'm only at about 70% commitment right now

It is not my fault the Project is not giving out work

:Pokes:You all need to write to Eon and voice your objection there:Pokes:
I was going to add 6 more nodes tonight That should add another 2% to my demand calls

The Admins need to upgrade the eon servers
They can not even handle the load from 1:hiya: guy working his Pharm out of just his house.
That is Sad if you think about it.

[NGS]Cpt00Kirk
09-09-2008, 02:24 AM
Yes ... it is Lauren2U ( however you not the "normal" house cruncher like the rest :rock:

ill send a mail to Eon .... its sucks like this! EDIT : mail send

LAURENU2
09-13-2008, 07:07 PM
ill send a mail to Eon .... its sucks like this! EDIT : mail send

I to have sent them mail But never once did I get a reply:cry:
So I thought I would get there attention in other ways:bathroom:
I Added more nodes today:thumbs:
Got about 80% of my Nodes back online now :cheers:
As the temp drops here I will even add more power
But I think I will have to use a little bit of my 2nd DSL line bandwidth
To coup with the added load
My one Linsys DSL router for the Pharm chokes at times under extreme load

Man this project sucks bandwidth :bang:

alpha
09-17-2008, 08:36 AM
Seconded! News and work units would be very nice. Maybe if enough of us poke the admins we'll get an answer. :Pokes:

Quoting myself here on purpose. I did send the admins an email when I posted this and got a prompt response from Graeme, but I forgot to mention it here. Received on 5th August 2008:


I've had a postdoc leave and a new student is taking over the project. He is getting up to speed, and we're going to start into some new calculations looking at grain boundaries in metal alloys. These will be more expensive calculations than we have been running, and yes, there will be lots of work units to come.
I will post an update very soon. My postdoc has done some very nice work with eon to improve our algorithm. His paper is almost through the review process; it's posted on our wiki in the 'Under Review' section:

http://theory.cm.utexas.edu/wiki/index.php/Manuscripts
L. Xu and G. Henkelman, Adaptive kinetic Monte Carlo for first-principles accelerated dynamics, JCP (2008).

Chuck
09-18-2008, 02:16 PM
:rotfl:Now You are the one talking too much, bring them on.... :fight:


Anyway gang.... there is a rather simple fix for those who wish to run Eon... GRANTED, there are still the issues at the server, but the clients (and WINDOWS mostly) are the problem.


If someone can turn this into a .reg file for those not willing to risk it..... Here is the WHAT IS WRONG and HOW TO FIX IT (client side).

Windows waits after closing a socket (to eon & others) before opening a new one. Due to the short job time with EON, we trip over both the server AND MOSTLY the client OS... not the client.

So, as referenced above about the 'tweaker' and the setting being put to 30... here is the key info.


start->run->regedit

now navigate your way down to:


\HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Paramters

(left side)...


On the right, you need to create a NEW (or change) a DWORD entry. So, right side 1/2 window, create a new key (or double-click on the existing) name "TcpTimedWaitDelay" of type "DWORD". DO NOT INCLUDE THE QUOTES

if you cycle through sockets REALLY fast, you may want 5-10 (respresenting seconds) after the program closes the socket (.... seeeee windows does a MANDITORY WAIT)... this parameter tells the driver how long to wait -or- use the default value (240-300 sec). Normal folk will want 20-30 seconds... I personally use 30 seconds and am fine..

you may want a shorter delay in closing a socket, perhaps 10 or 15. Adjust accordingly to your needs.

The change(s) take effect ONLY AFTER a reboot. The TCPIP driver reads it at startup only.

This will also effect any BT traffic you might be sharing with a friend.

Windows machines used as project servers usually have about 128 effective sockets before they go nuts.... (ignore that 600-800 crap)... So based on 128.... setting the delay value == 5 seconds ensures all data gets there and the 'ack' is received in the final close.... also, your windows kernel doesn't go nuts having more than the default 128 sockets it allocates. :)


Enjoy all.... So far, it has helped a few projects very nicely.



And in the words of one that came before me....''' Wash, Rinse, Repeat..... '''..


but here it is. It works on all I have , but not verified on Vista (since it's BSD underneath). Perhaps a Vista owner will augment this and help out the rest, but win/2k HAULS BUTT...

You know you are doing good when a 10 Mbit DSL sustains 1.2 MB/sec download. (yes, synchronous clocked)


C.


PS: YES, I am back... Not sure how long or how much, but at least in limited fashion.


PPS: The 'daring' may use this as a reference
http://publib.boulder.ibm.com/wasce/V2.0.0/en/tuning-windows.html

LAURENU2
09-18-2008, 04:36 PM
I see the Dutch Power Cows are making a move they will be at our back in 22,138.80 Days :lmao:
They exceeded our output by 55 points so far for to day
:idea:I guess I need to get more Q's back online:rock:

em99010pepe
09-18-2008, 04:37 PM
Better move my Cyrix 166 MHz to eOn.

IronBits
09-18-2008, 07:40 PM
TcpTimedWaitDelay.reg (Windows XP)

Windows Registry Editor Version 5.00

[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters]
"TcpTimedWaitDelay"=dword:0000001e


Mine was already like that Chuck :)

Chuck
09-18-2008, 08:11 PM
IB

Any good tuning program will put that in. Hex 1e is, of course, 30 decimal. feel free to change 0000001e to a decimal value and lower if you need it. (the leading zeros indicate hex format).... anything else is presumed decimal.

and where did this file come from because it's a file XP, 2K, ME, 98 and I think 95 will read directly into the registry -> TcpTimedWaitDelay.reg (Windows XP). This looks like you have a 'Windows XP' version version of a delay param.


Chuck

Chuck
09-18-2008, 08:25 PM
I see the Dutch Power Cows are making a move they will be at our back in 22,138.80 Days :lmao:
They exceeded our output by 55 points so far for to day
:idea:I guess I need to get more Q's back online:rock:



lauren,
edit your registries... make it a REG4 format file (look it up), extension .reg, and let each machine of the pharm load it. that will make eon go faster and 'warm up' your linksys :)

C.

LAURENU2
09-18-2008, 09:44 PM
Better move my Cyrix 166 MHz to eOn.
:cheers::thumbs:


Chuck Reg Files scare the Bytes out of me :scared:
Besides I am To Dumb to play with Reg files Ask IB he can vouch for that :looney:
I just like :beep: Fast

I am Bring back more nodes little by little as the weather gets cooler here
Added some more today to compensate for the Cow's increase in output

Sorry That my computer room upgrade took far longer then I had planed
It has been a vary hard summer for me :cry:

Chuck
09-18-2008, 10:02 PM
Lauren,

all understood.

would you feel better if I were to prepare a proper REG file for you? I know it will help. you can try it if you like and if you don't i have a way of removng it. NO DAMAGE TO SYSTEM. Interested??? We would arrange for private exchange to prevent tampering. (so you cna blame me and only me if it messes up... lol)


re the cows: They are and have been turning the fields here already.... BOY IS THE FERTILIZER GOOD AND LIQUID!!!!!

Like you, I am adding more hardware as the days cool. here is 60 @ night, 72+ during the day (and my office is on the WEST side of the house... not to smart @ 80+ ambient, eh?)

I'll be back up to full capacity shortly.


Have a new build in progress.

Quad CPU , K-10 Quad Cores. Tyan 2915 mobo. That will allow the 252's, etc to retire to my son or whoever needs them. I MIGHT make a disk farm out of 1... Sata-2, + E-sata + SAS... It would just fit. That + the new TB drives and I can fit a lot into 1 box and a 2U daughter box (total 6U height). The mobo cost isn't bad, but best cpu price I can find is $2100 a pop... OUCH & DOUBLE OUCH!

Suggestions????



Thoughts???


C.

LAURENU2
09-18-2008, 10:19 PM
I would just fudge it up :bang: I know it :bang: But if IB Feels it is worth it he has the Keys to my network.
But it might be vary slow going on that network with all the Nodes Running EoN
I even add some Eon clients on this DSL line

Chuck
09-18-2008, 10:31 PM
Please forgive me, but given the class pointers, i suggest caution. IB, please correct me if all those registry entries are needed or you had multiple connections at one time?

It is clear I don't understand Reg5 format, for which I ask forgiveness. MS pulled another cute move... you only need that ONE parameter and it does have to (regardless of version) align with the class-ID else it will not take effect. That is why I put it under 'current control set'.

Please forgive my living in Linux. WIndows seems to be way to unmanagable and understandable now.

With thanks and respect,
Chuck




Actually, to use it below the XP version of windows, including w2k, you will have to change the 5.00 to 4.00
I exported it from my registry, then removed all the other 'stuff'

LaurenU2, save it, remove the .txt at the end, then double click it http://www.free-dc.org/forum/images/smilies/smile.gif

Chuck
09-18-2008, 10:35 PM
At least I do understand the 5 vs 4...


that's the NT-5 vs NT-4 kernel. :)


all those service packs and 'security changes' are what scare me.

All I respectfully ask, is a checkpoint be made BEFORE adding it so it can be undone easily.

thanks Gents,
C

IronBits
09-18-2008, 10:35 PM
Good catch Chuck!!!!!!!!!!!!! Holy crap, that was so totally stupid of me to post an exported reg entry without removing all the computer specific crap in there.
LaurenU2, you don't need it, you already take too much work away from everyone else, I'm not gonna help you take it ALL away on EON :p

Chuck
09-18-2008, 10:41 PM
IB,
would you be kind enough, using my posting as a model, to set a baseline for all of FreeDC to use.... then send a slightly SLOWER one to Lauren??? :)

It appears you have, from your XP template, all the info needed to transform my data to your Reg5 format and still keep it a 1 line entry for the TCPIP driver.


With thanks,
Chuck

IronBits
09-18-2008, 11:07 PM
That's a great idea! :D

Here is the edited version, just rename it to TcpTimedWaitDelay.reg and double click it to set the delay to 30 as you suggested :)

Chuck
09-18-2008, 11:59 PM
...... <grinning>.....


Just remember boys and girls... you have to REBOOT windows AGAIN (like you didn't already know that... LOL) AFTER you add the registry entry to your existing system.

The presence of and value (if any) is read during TCPIP driver init @ kernel startup. It is never examined after that as it sets the driver config for the whole machine.


Since each user has different needs, it's ok to edit the value of the key as needed... do NOT set == 0 (0== indefinite wait time.. which translates to 10 minutes or so in the NT5/XP kernel)

It is very much like the linux 'select()' call or 'poll()'.... when the time delay is '0' that means wait forever, whereas '(char *) 0' ...aka... NULL, means DO NOT wait.

Vista owners beware....such issues could and most likely are lurking in your systems. Go slow and be careful. ..... Or get a real OS!!!!! :rotfl:


...ooops, did i say that outloud??? sorry!!!

C.

LAURENU2
09-19-2008, 12:42 AM
LaurenU2, you don't need it, you already take too much work away from everyone else, I'm not gonna help you take it ALL away on EON :p
But:cry: But :cry:I was just doing it for the Team
em99010pepe made me do it:rock:

IronBits
09-19-2008, 02:02 AM
:bigtrain: :thumbs: :smoking:

LAURENU2
09-19-2008, 09:38 AM
I added 2 more nodes last night and it seemed to do the trick
Now I am back sucking up all the spare WU's Again :lmao:

Paratima
09-19-2008, 11:02 AM
We noticed! :bang:

LAURENU2
09-19-2008, 05:30 PM
We noticed! :bang:
Now if you were the EoN's system Adim I would be happy
At about 88 % Right now and still climbing Rack by Rack

Chuck
09-19-2008, 05:52 PM
Pssssttttt...


Ok gang, Lauren is happy now!!!

Let's gets some real work done!


But Lauren, did that Reg adjustment do the job for you?


:)

:cheers:

LAURENU2
09-19-2008, 07:34 PM
Pssssttttt...


Ok gang, Lauren is happy now!!!

Let's gets some real work done!


But Lauren, did that Reg adjustment do the job for you?


:)

:cheers:
I don't do Regs They scare the Bytes out of me:rock: I told you
I just push harder with more Ghz
"happy now " hahahaha Ya I see a Q8500 combo for $199.
:umm:and I have a Empty slot in one of my Racks That JUST NEEDS to be filled

So I :Pokes:guess :Pokes:I :Pokes: will Just have to go out tonight and get it

em99010pepe
09-19-2008, 08:42 PM
EON client controller by DPC

http://www.pubquizcentre.nl/DigiK-oz/dceon.aspx

IronBits
09-19-2008, 09:19 PM
Here's how I do it :)

@echo off

goto 6

:8
for %%i in (1 2 3 4 5 6 7 8) do CALL :RUNIT %%i
goto :END

:6
for %%i in (1 2 3 4 5 6) do CALL :RUNIT %%i
goto :END

:4
for %%i in (1 2 3 4) do CALL :RUNIT %%i
goto :END

:3
for %%i in (1 2 3) do CALL :RUNIT %%i
goto :END

:2
for %%i in (1 2) do CALL :RUNIT %%i
goto :END

:RUNIT

D:
cd \dcprojects\EON\eon%1 & start /min client.exe

goto :EOF

:END

:EOF

LAURENU2
09-20-2008, 08:25 PM
EON client controller by DPC

http://www.pubquizcentre.nl/DigiK-oz/dceon.aspx
I did what they said
But I am not sure it worked
I set it for 4 started it . I did not see any windows or CPU usage, However I did check Task-manager and I did see 4 client.exe's running.
I got worried they would do nothing so I went back to the old way

Is this the way it works or was something wrong

LAURENU2
09-21-2008, 01:24 AM
antarax== Alliance Francophone :sniper:==12,842,959 in one day NO WAY :bang:

stalin12
09-21-2008, 05:27 AM
Windows waits after closing a socket (to eon & others) before opening a new one. Due to the short job time with EON, we trip over both the server AND MOSTLY the client OS... not the client.

antarax
09-22-2008, 05:45 PM
antarax== Alliance Francophone :sniper:==12,842,959 in one day NO WAY :bang:



Hello, just because of change of email address, the other going to be deleted ^^ :jester:

em99010pepe
09-22-2008, 05:53 PM
antarax,

Do you usually go to see the "24 Heures du Mans"?

Carlos

antarax
09-22-2008, 06:45 PM
yes I'm going to the 24 Hours of Le Mans, and also testing :)

I am not very far from the circuit :thumbs:

em99010pepe
09-23-2008, 05:41 AM
Testing?

antarax
09-23-2008, 07:22 AM
yes before the start of the 24 Hours of Le Mans, 2 evenings ago for tests on Wednesday, and Friday and I look at it is testing :)

http://www.lemans.org/accueil/index_gb.html

LAURENU2
11-25-2008, 01:21 AM
:confused:Is this the end of EoN as we now know it. Or is this just a hardware failure :Pokes: Anyone?

LAURENU2
11-28-2008, 11:09 AM
:confused:Is this the end of EoN as we now know it. Or is this just a hardware failure :Pokes: Anyone?
I guess no one knows what is going on with EoN

LAURENU2
12-01-2008, 06:40 PM
EoN is Back going at full speed :roadkill:

Chuck
12-01-2008, 11:59 PM
Yep, caught it and got a few hundred up there already.... saw it about 2 hours ago.... not sure what today's stats will be.... also only have two machines flipped over from D.net.... Need to be graceful. Tomorrow things will be full speed and the queues will be full! :) SOMEONE is hogging the WU's :rock:


It's :roadkill: time!!!!

C.

Xaverius
12-02-2008, 06:35 AM
I did what they said
But I am not sure it worked
I set it for 4 started it . I did not see any windows or CPU usage, However I did check Task-manager and I did see 4 client.exe's running.
I got worried they would do nothing so I went back to the old way

Is this the way it works or was something wrong

Did you install it in a new directory are an used one?
I usually start as many clients as I have core's in my computer (that's 2) and the proxybuffer for about 15 minutes, not much but normally enough.

It can take some time to get the work (normally not more than a few minutes) but you can see "requestors running" and "requests queued" in the program.

You did specify the clients to connect to the proxy? :D

LAURENU2
12-02-2008, 10:13 AM
Did you install it in a new directory are an used one?
:D
No I am not using the client controller :blush: I guess I am just to stupid :gone: to know how to use it :cry:

Death
12-02-2008, 10:52 AM
can you just follow the instructions on a client controller page? ))))))))

this is my 666-th post!

Xaverius
12-02-2008, 12:44 PM
I guess English is sometimes a bit hard to understand, isn't it. :looney:
I also have that problem, most of the time when it's about 4-5 am and the bloodlevel in my alcohol.... oh...., whatever. :blues:

:D

LAURENU2
12-03-2008, 09:30 AM
I can not read is not the only problem I have here.
I will have to download well over 3 terabytes of upgrades to my OS's just to get that program to work.
If I started now given the time I have at hand I might be able to do 2 a day if every thing went as planed
And that would bring me well into the New Year.:cry:

Death
12-03-2008, 10:08 AM
Laurenu2 you mean all updates from microsoft update? do you really mean tera? not peta?
cuz you can download xp - it's only near one Gb )))))

LAURENU2
12-03-2008, 05:28 PM
Laurenu2 you mean all updates from microsoft update? do you really mean tera? not peta?
cuz you can download xp - it's only near one Gb )))))
:clap: I mean 3300 Megabytes
Net framework is 55MB X 60+ nodes =3.3 terabyte I thought :dunno:

alpha
12-04-2008, 02:25 AM
55MB * 60 is 3.3 gigabytes.

You'd need to have nearly 63,000 nodes for it to be 3.3 terabytes.

Death
12-04-2008, 03:42 AM
Laurenu2, oh man, why did you download it 60 times? why don't just download offline install one time and install it everywhere? or tell your proxy to cache larger files, so all nodes just take it from cache?

Chuck, now I really understand your tagline about professor. )))) really. make a lecture in answer to short joke - this is wonderful.
Ok, my first HDD was 40 Mb - full 41943040 bytes. I was very happy to have almost 2 megabytes for "free" (no, no, not giga, MEGA). So I know a little about mega and giga. ))))))
And I really loves to see an overall scitech boost. Havin' a 32Gb flash drive - I'm amazed that its almost 1000 times larger that my first hdd and simply fits in a pocket.

thats marketing simplification - measure hdd size in a mega (in a meaning of millions, not 2 powers) looks stupid to me. one buy 500 Mb HDD, plug it, and in a properties tab windows tell him "this is 486 Mb capacity". and on most drives that I seen, there was a clarification on label - kinda like "we say that 1 gb means 1 billon bytes, but real capacity is 976 mb" or something like that.

Death
12-04-2008, 03:49 AM
Laurenu read alpha's post.

I'm trying to tell you the same.

kilo
mega
giga <--- you miss this. ))))))
tera
peta

LAURENU2
12-04-2008, 09:13 AM
Laurenu read alpha's post.

I'm trying to tell you the same.

kilo
mega
giga <--- you miss this. ))))))
tera
peta
See I told you I was stupid :scratch: And here you think I have the smarts to mass distribute a update to all nodes ,
or point all nodes to one proxy server :lmao: When I can not even count to 1000 :rotfl:
PS I have even have a hard time getting my wife to work right :slap:

gopher_yarrowzoo
12-04-2008, 11:15 AM
Yes well that and the fact MS don't make it exactly easy for you to download offline update patches..

Paratima
12-04-2008, 10:38 PM
How about a computer with 4K of RAM ??? Yes, 4K.... and that made you 'The Man' !!!!

*SOMEWHERE* I have my FULL HEIGHT 10MB IBM PC 5 1/4" drive. I also have a 10MB FULL SIZE platter and a 60MB 'stack' (from the CDC drives that were the size of a washing machine).
Been there; done that! :D Hell, I programmed electro-mechanical micro-computer accounting machines with 400 frigging WORDS (12 decimal digits/word).
Off-line storage? Yep: paper tape! :rock: The mighty Burroughs E-4000! Un be liev a ble.
Like trying to get a quart of fresh frogs in a pint pickle pot. :eek:

IronBits
12-05-2008, 07:40 AM
EON is not worth all that :bang:

LAURENU2
12-05-2008, 09:56 AM
EON is not worth all that :bang:
:confused: Can you elaborate a little more on that IB :Pokes:

IronBits
12-05-2008, 05:44 PM
It's simple...
Not enough work to go around... client server communications are less than desireable...
Admins are not very responsive, server goes down a lot...
Folks are forced to find ways to take more work than they should to maintain higher output...
One such way is to use a proxy program that sucks the life out of the server so no one else can get work...
Said programs require 55mb of .net add-ons to windows x 60 boxen (in your case)

...
...
...

LAURENU2
12-05-2008, 11:03 PM
OH OK
I thought you were implying the project focus was not worth our efforts
As for project management and the WU supply chain might be due to a funding problem.:cry:
As for doing the update network wide I have decided to keep it to only 10 nodes :thumbs:

Xaverius
12-06-2008, 06:19 PM
I can not read is not the only problem I have here.
I will have to download well over 3 terabytes of upgrades to my OS's just to get that program to work.
If I started now given the time I have at hand I might be able to do 2 a day if every thing went as planed
And that would bring me well into the New Year.:cry:I probably bang my head afterwards if I give yout this information, but.... You just need to download the update once and run the proxy on that only computer. Then the only thing is to point all the other installations of eon to the IP of that computer. One proxy could provide all the work you need for all your boxes.

That is how I did it. I ran eonstart on my XP-installation and I edited the config on my Kunbuntu to 192.168.1.103 to get work. (yes, I only have two PC's :|party|:)


It's simple...
One such way is to use a proxy program that sucks the life out of the server so no one else can get work...
Said programs require 55mb of .net add-ons to windows x 60 boxen (in your case)
...Perhaps we don't actually understand eachother but what't the difference of having 8 installations per core (nice on a octacore) hammering the server and one proxy with let's say 20 requestors for work for an entirely netwerk of eon-installations.

You're being quite negative about the proxy but it have you actually tried it for a while (no, not just for a day) and see how it works??? If enough work is buffered most (or even all) of the requestors are in a "pause"-mode and do not request even more work.

And even if it is hammering the server the way as 8 clients per core does, how come you are negative only to this proxy program and not to those people who are installing 8-installations per core... Even LAUREN2 admits he's trying to hammering the server so hard that the admins have to upgrade it...

Or am I missing some carcasm somewhere? I'm onlu Dutch you know. :cow:

LAURENU2
12-06-2008, 11:33 PM
not to those people who are installing 8-installations per core... Even LAUREN2 admits he's trying to hammering the server so hard that the admins have to upgrade it...


Ouch 8 per core :dunno: I was only doing 12 to 14 per quad node


As for trying to hammering the server so hard :lmao:

I do that to every project I work :smoking:

It was just that EoN was one of the few that stalled out under the stress

Paratima
12-11-2008, 09:39 PM
http://lesgphoto.com/images/deadhorse2.gif The thread that would not die! :gone: :lmao:

LAURENU2
12-11-2008, 11:43 PM
http://lesgphoto.com/images/deadhorse2.gif The thread that would not die! :gone: :lmao:

I sure hope that was a mare http://lesgphoto.com/images/deadhorse2.gif :lmao:

Xaverius
12-17-2008, 05:16 PM
Seems the server is up againg since a few days (a little staying alive bump):moto:

LAURENU2
12-21-2008, 01:41 PM
Well Enjoy the extra WU's As :hiya:I turned down the Pressure in EoN to make a massive DUMP :dump: in DPAD
:umm:Get your WU's now before I power back up:rock: It won't be long

LAURENU2
01-01-2009, 06:34 PM
http://lesgphoto.com/images/deadhorse2.gif The thread that would not die! :gone: :lmao:

Yippy I just took 5th Place :roadkill:
:cheers::banana::cheers:

Bok
01-01-2009, 07:37 PM
sweet!

I noticed you were getting close!

:allhail:

Chuck
01-05-2009, 01:22 AM
Gang, we seem to have two Eon teams....or two EON projects (?)


Our own stats show it.... I sent a PM to a few admins.... but ask how to get our association to the PROPER 'Free-DC' team... something has changed and would appreciate some assistance. A bunch of users show up as NEW Eon users with NO TEAM NAME on the team page. Lauren doesn't even show up!!!! He, like I, am "archived".

Lauren.... you're about to get some SERIOUS heat!!!!! You KNOW I can steal them from you..... and now I have HELP!!! Care to get the 'archived' vs 'active' situation corrected and then let's have SOME SERIOUS FUN as well as take 4th place ?????? Let's see how long we can keep ALL the EON WUs in FDC posession!!! ??

Oh, and Fyi..... I can save MORE than 10000 WUs at a time WITHOUT them expiring :rock::roadkill:... AND distribute them across the Linux and Windows sides.... :)

------------ To more serious matters -------------------------

EDIT: Found this doing an in depth search on the Eon server...... How does one fix it? It obviously occurred after the 1st but before today (where the 'eon we know' is 'archived' on OUR server.... when it really isn't)




Warning: mysql_num_rows(): supplied argument is not a valid MySQL result resource in /home/eon/public_html/stats.php on line 320

No such user was found in the team Free-DC
Warning: mysql_fetch_array(): supplied argument is not a valid MySQL result resource in /home/eon/public_html/stats.php on line 322

Total number of results: 594410640




Thanks,
C

*EDIT* by Bok. Just a misunderstanding and using wrong parameters on one of the pages. Explained in a PM that the page requires a &team=Free-DC parameter in the url.. no harm, no foul :)

Bok
01-05-2009, 09:31 AM
Chuck, without providing any examples of what you mean it's impossible to answer this?

Eon archived ? Why, how do you think this? It's not in the archived section (http://stats.free-dc.org/stats.php?page=index)

I've looked through the Free-DC stats and can see nothing wrong..

There is only one Free-DC team in the teamlist (http://stats.free-dc.org/stats.php?page=teams&proj=eon) and that's confirmed in the DB and the export that Eon provide.

Lauren still has 18M+ points.

So where are you looking ? You mention Eon archived on Our Server. Who's server are you talking about here ?

As for mysql errors, well they basically occur when an invalid query is made for whatever reason.

I'm just totally confused, not that that's difficult these days :)

Chuck
01-05-2009, 12:48 PM
Error found!
PILOT ERROR!



Chuck, without providing any examples of what you mean it's impossible to answer this?

Eon archived ? Why, how do you think this? It's not in the archived section (http://stats.free-dc.org/stats.php?page=index)

I've looked through the Free-DC stats and can see nothing wrong..

There is only one Free-DC team in the teamlist (http://stats.free-dc.org/stats.php?page=teams&proj=eon) and that's confirmed in the DB and the export that Eon provide.

Lauren still has 18M+ points.

So where are you looking ? You mention Eon archived on Our Server. Who's server are you talking about here ?

As for mysql errors, well they basically occur when an invalid query is made for whatever reason.

I'm just totally confused, not that that's difficult these days :)

em99010pepe
01-05-2009, 12:56 PM
Target locked....going down....

Chuck
01-05-2009, 01:29 PM
Target locked....going down....


BULLS***!!! :D I've side-slipped 165,000+ lbs before in a cross-wind.... I am NOT GOING DOWN..... may just have to make another pass is all.... but I WILL do a 3-point touchdown!!!

C

LAURENU2
02-08-2009, 04:03 PM
:cow: starting a stampede here in EON
They have increased there output day by day over the past
4 days and are now topping our output.:cry:
We are not on there Radar Yet, :umm:But Team HD20 in Only has
16,354 Days :lmao:before Dutch Power Cows take them Down

I could increase my output to 2X what it is now but it seems
All I do is Bog down there server and total output drops :cry:

IronBits
02-08-2009, 05:06 PM
It's not a project we can do anything about, so I no longer pay any attention to EON.
There is not enough work to go around, and their servers can not handle the load.
If you like the project, then enjoy yourself, else, find work elsewhere in another project.