Tom (Dec 23 2008)


log in

Advanced search

Message boards : Technical News : Tom (Dec 23 2008)

1 · 2 · 3 · Next
Author Message
Profile Matt Lebofsky
Volunteer moderator
Project administrator
Project developer
Project scientist
Avatar
Send message
Joined: 1 Mar 99
Posts: 1384
Credit: 74,079
RAC: 0
United States
Message 844300 - Posted: 23 Dec 2008, 23:00:32 UTC

Today had our weekly outage for mysql database backup, maintenance, etc. This week we are recreating the replica database from scratch using the dump from the master. This is to ensure that the crash last week didn't leave any secret lingering corruption. That's all happening now as I type this and the project is revving back up to speed.

Had a conference call with our Overland Storage connections to clean up a couple cosmetic issues with their new beta server. That's been working well and is already half full of raw data. Once the splitters start acting on those files the other raw data storage server will breathe a major sigh of relief. I was also set to (finally) bump up the workunit storage space yesterday using their new expansion unit - but waited until their procedure confirmation today lest I did anything silly and blew away millions of workunit files by accident. The good news is that I increased this storage by almost a terabyte today, with more to come. We have officially broken that dam.

I also noticed this morning the high load on bruno (the upload server) may be partially due to an old, old cronjob that checks "last upload" time and alert us accordingly. This process was mounting the upload directories over NFS and doing long directory listings, etc. which might have been slowing down that filesystem in general from time to time. I cleaned all that up - we'll see if it has any positive effect.

Jeff's been hard at work on the NTPCker. It's actually chewing on the beta database now in test mode. We did find that an "order by" clause in the code was causing the informix database engine to lock out all other queries. This may have been the problem we've been experiencing at random over the past months. Maybe informix needs more scratch space to do these sorts, and it locks the database in some kind of internal management panic if it can't find enough. Something to add to the list of "things to address in the new year."

- Matt

____________
-- BOINC/SETI@home network/web/science/development person
-- "Any idiot can have a good idea. What is hard is to do it." - Jeanne-Claude

Profile Bernie Vine
Volunteer tester
Avatar
Send message
Joined: 26 May 99
Posts: 6610
Credit: 22,431,971
RAC: 14,476
United Kingdom
Message 844306 - Posted: 23 Dec 2008, 23:08:26 UTC

Thanks Matt

One of the first things I do each day is to check the Technical News thread.

Thank you for keeping us all up to date on the inner workings of SETI@Home

May I wish you a Good Christmas and New Year.

Bernie
____________


Today is life, the only life we're sure of. Make the most of today.

Profile Blurf
Volunteer tester
Send message
Joined: 2 Sep 06
Posts: 7266
Credit: 6,273,118
RAC: 2,310
United States
Message 844319 - Posted: 23 Dec 2008, 23:17:46 UTC

Thx Matt-enjoy your holiday!
____________


Profile Dr. C.E.T.I.
Avatar
Send message
Joined: 29 Feb 00
Posts: 15988
Credit: 683,364
RAC: 101
United States
Message 844367 - Posted: 24 Dec 2008, 1:26:15 UTC


. . . Thanks Matt - Good to hear something regarding NTPCker too

< All @ Berkeley have a Wonderful Christmas Eve - Enjoy the Holidays


____________
BOINC Wiki . . .

Science Status Page . . .

Profile SATAN
Avatar
Send message
Joined: 27 Aug 06
Posts: 835
Credit: 2,094,367
RAC: 0
United Kingdom
Message 844370 - Posted: 24 Dec 2008, 1:28:49 UTC

Happy holidays guys. Thanks for some very welcome good news Matt.
____________

Speedy
Volunteer tester
Avatar
Send message
Joined: 26 Jun 04
Posts: 623
Credit: 4,780,174
RAC: 5,509
New Zealand
Message 844383 - Posted: 24 Dec 2008, 1:52:55 UTC
Last modified: 24 Dec 2008, 2:00:27 UTC

Thanks for the updates Matt. Hope you have a nice work free break. How long are the hard working crew at the lab taking off?

Seasons greetings to all
____________

Live in NZ y not join Smile City?

Profile Ace Casino
Avatar
Send message
Joined: 5 Feb 03
Posts: 283
Credit: 18,484,583
RAC: 19,041
United States
Message 844531 - Posted: 24 Dec 2008, 11:55:57 UTC

If any of the problems such as upload, running out of WU’s, bandwidth limits has anything to do with CUDA being released, why not limit the number of CUDA WU’s you can request per day to a lower number, until thing sort themselves out.

Only allow 20, 40, 60 WU’s a day per computer.

I don’t know if this is possible, you only allow a certain number of Wu’s per CPU on a computer. Why not only allow a certain number of WU’s per NIVIDIA card? Something LOW for now, then turn it up slowly over time.

Happy Holidays

George
Volunteer tester
Send message
Joined: 14 Oct 08
Posts: 100
Credit: 435,680
RAC: 0
United States
Message 844542 - Posted: 24 Dec 2008, 12:59:30 UTC

they are limited to 100 wu's per computer cpu core same as all others just they can do that much work where as cpus can't

PhilG57
Send message
Joined: 18 Nov 03
Posts: 13
Credit: 3,935,892
RAC: 1,455
United States
Message 844543 - Posted: 24 Dec 2008, 13:00:11 UTC

Matt - I read and appreciate the technical updates. I can only imagine the hundres of problems, large and small, you and your team handle every day. But I'm confised: I read the project is coming back up, see there are 55,818 workunits to be downloaded, but when I request more workunits, the status returned is 'no work available'. What's up? THX.
____________

PhilG57
Send message
Joined: 18 Nov 03
Posts: 13
Credit: 3,935,892
RAC: 1,455
United States
Message 844545 - Posted: 24 Dec 2008, 13:03:57 UTC - in response to Message 844543.

Hmmm. Now, a couple of minutes later, it's found and downloaded some work for me. I'm good for awhile. THX.
____________

Profile Matt Lebofsky
Volunteer moderator
Project administrator
Project developer
Project scientist
Avatar
Send message
Joined: 1 Mar 99
Posts: 1384
Credit: 74,079
RAC: 0
United States
Message 844631 - Posted: 24 Dec 2008, 17:26:11 UTC - in response to Message 844543.

I read the project is coming back up, see there are 55,818 workunits to be downloaded, but when I request more workunits, the status returned is 'no work available'. What's up? THX.


What's probably happening here is that, yes, there is work "available." However, when your client requests work from our scheduling server, the scheduler process looks at the "feeder" which holds at any given time the names of 100 available workunits to send out. So the feeder process has to constantly fill its tiny cache, and to do so queries the database every two seconds to see if there's more work available. Long after the project comes back up the database is quite overloaded, so it may not respond very fast. In fact, it sometimes takes many minutes for it to cough up results to the feeder, during which clients get "no work available."

In other words, the feeder is like a single cashier in a large department store. Sometime the cashier needs to make change, which holds up the entire line, even though there's plenty of money kept elsewhere behind the counter.

- Matt
____________
-- BOINC/SETI@home network/web/science/development person
-- "Any idiot can have a good idea. What is hard is to do it." - Jeanne-Claude

Profile Dr. C.E.T.I.
Avatar
Send message
Joined: 29 Feb 00
Posts: 15988
Credit: 683,364
RAC: 101
United States
Message 844637 - Posted: 24 Dec 2008, 17:42:44 UTC - in response to Message 844631.

I read the project is coming back up, see there are 55,818 workunits to be downloaded, but when I request more workunits, the status returned is 'no work available'. What's up? THX.


What's probably happening here is that, yes, there is work "available." However, when your client requests work from our scheduling server, the scheduler process looks at the "feeder" which holds at any given time the names of 100 available workunits to send out. So the feeder process has to constantly fill its tiny cache, and to do so queries the database every two seconds to see if there's more work available. Long after the project comes back up the database is quite overloaded, so it may not respond very fast. In fact, it sometimes takes many minutes for it to cough up results to the feeder, during which clients get "no work available."

In other words, the feeder is like a single cashier in a large department store. Sometime the cashier needs to make change, which holds up the entire line, even though there's plenty of money kept elsewhere behind the counter.

- Matt



. . . now that's clear and to the Point - Thanks for the Update and have a Good Holiday Sir!


____________
BOINC Wiki . . .

Science Status Page . . .

Eddy
Send message
Joined: 6 Jun 00
Posts: 3
Credit: 33,079
RAC: 0
Australia
Message 844959 - Posted: 25 Dec 2008, 11:07:03 UTC

quick question, can you have more than two projects running at the same time?

Profile littlegreenmanfrommars
Volunteer tester
Avatar
Send message
Joined: 28 Jan 06
Posts: 1410
Credit: 934,158
RAC: 0
Australia
Message 844962 - Posted: 25 Dec 2008, 11:46:42 UTC - in response to Message 844959.

Quick answer: YES
____________

Profile Jason Tobin
Avatar
Send message
Joined: 28 Apr 07
Posts: 21
Credit: 1,014,494
RAC: 0
Canada
Message 845171 - Posted: 26 Dec 2008, 0:45:57 UTC - in response to Message 844531.

Hi;

It seems I am not experiencing any faster computation speed from the latest Boinc Client 6.4.5 using the
CUDA add on. My project is Seti@Home

OS: Windows XP Media Center (Pro) 32 bit w/ Sp3

Video Card: NVidia GeForce 8800 GT

Intel Core2 Quad CPU Q6600 @ 2.4 GHz
2.0 GB RAM

driver version: 180.48

I down loaded the the new Boinc client.

Downloaded the latest NVidia Driver (ver.180.48)

Restarted Client stated that it did have CUDA Compatible components.
The option for the GPU is enabled on my settings.
But I am not seeing any benefit.
I am not getting any error messages what so ever.

Any thoughts?? AM I doing something wrong?

Jason Tobin

jasontobin48@hotmail.com
____________
Jason Tobin
Alien Hunting Specialist

Speedy
Volunteer tester
Avatar
Send message
Joined: 26 Jun 04
Posts: 623
Credit: 4,780,174
RAC: 5,509
New Zealand
Message 845176 - Posted: 26 Dec 2008, 1:13:57 UTC

It seems I am not experiencing any faster computation speed from the latest Boinc Client 6.4.5 using the
CUDA add on. My project is Seti@Home

Have you read this thread? Hopefully this will help
____________

Live in NZ y not join Smile City?

Eddy
Send message
Joined: 6 Jun 00
Posts: 3
Credit: 33,079
RAC: 0
Australia
Message 845545 - Posted: 27 Dec 2008, 4:46:02 UTC - in response to Message 844962.

Quick answer: YES

so how do you do it? long answer please

OzzFan
Volunteer tester
Avatar
Send message
Joined: 9 Apr 02
Posts: 13307
Credit: 27,894,957
RAC: 16,306
United States
Message 845577 - Posted: 27 Dec 2008, 6:45:09 UTC - in response to Message 845545.

Quick answer: YES

so how do you do it? long answer please


First, it depends on what you mean by "projects". If by "projects" you mean tasks or workunits, then you need a CPU (logical, virtual or real) for each task you want to run. For instance, if you have a dual core CPU, or a dual CPU machine, or a CPU with Intel Hyperthreading, you can run at most two tasks. If you have a quad core CPU, or a quad CPU system, or a dual core with Intel Hyperthreading, you can run at most four tasks at once. And so on and so forth.

If by "projects" you mean different BOINC projects, you must first attach your computer to additional projects to download their work, but note the same CPU limitations apply for CPUs in your machine as stated above (i.e. you cannot run more tasks than you have CPUs in your machine). Also, you cannot explicitly state which task or project runs on which CPU as this is handled by BOINC.
____________

Profile littlegreenmanfrommars
Volunteer tester
Avatar
Send message
Joined: 28 Jan 06
Posts: 1410
Credit: 934,158
RAC: 0
Australia
Message 845775 - Posted: 27 Dec 2008, 22:15:05 UTC - in response to Message 845577.

Thanks for stepping in there, Ozz... I had come back in, but missed Eddy's reply.

I should add, though, that since installing CUDA, I am presently running threee work units simultaneously.

(2 x SETI and 1 x SETI Beta)
____________

Eddy
Send message
Joined: 6 Jun 00
Posts: 3
Credit: 33,079
RAC: 0
Australia
Message 845854 - Posted: 28 Dec 2008, 4:02:01 UTC

thanks just like to help as much as possible, omly dual core PC so two tasks it is

Cheers

1 · 2 · 3 · Next

Message boards : Technical News : Tom (Dec 23 2008)

Copyright © 2014 University of California