Advanced search

Message boards : Server and website : Could we have a Server and WU situation update?

Author Message
Betting Slip
Send message
Joined: 5 Jan 09
Posts: 574
Credit: 1,911,306,275
RAC: 1,783,457
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 45664 - Posted: 12 Dec 2016 | 8:20:56 UTC
Last modified: 12 Dec 2016 | 8:48:18 UTC

The server download problems persist with stuck downloads and sometimes slow speeds. I would imagine these should be visible in your server logs.
It's also coming up to christmas so would be great to get that WU queue loaded up to take us through to the New Year.
Updates on the problems would be appreciated even as I know calls for updates are a pain in the arse.

EDIT to add:

You haven't even got "GPUGrid Preferences" and "Computing Preferences" pages sorted yet they have HTML errors and this really is "bread and butter" stuff. You must have a load of IT students within yards of you that would be willing to help.

We had, prior to server upgrade and WU shortage got up to 2.25 Petaflops in computational power, we're back dow to 1.5. Some may never return.

PappaLitto
Send message
Joined: 21 Mar 16
Posts: 240
Credit: 970,319,881
RAC: 3,543,850
Level
Glu
Scientific publications
watwat
Message 45668 - Posted: 12 Dec 2016 | 13:33:17 UTC

My suggestion to the scientists is even if there is no need for the extra computational power, make redundant WUs, in science, the more data the better. And this will keep the crunchers happy and stay with GPUGrid. Of course have some way to prioritize your not yet redundant WUs.

Profile caffeineyellow5
Avatar
Send message
Joined: 30 Jul 14
Posts: 225
Credit: 2,658,976,345
RAC: 0
Level
Phe
Scientific publications
watwatwatwatwat
Message 45675 - Posted: 12 Dec 2016 | 15:54:43 UTC - in response to Message 45668.

My suggestion to the scientists is even if there is no need for the extra computational power, make redundant WUs, in science, the more data the better. And this will keep the crunchers happy and stay with GPUGrid. Of course have some way to prioritize your not yet redundant WUs.
Been there. You are not gonna win that argument here on these forums. Too many people believe no work is better than 'busy' work, even if it means getting new and retaining old crunchers who see an empty project and turn around and walk away. Everybody claims, which they have a point on it, that BOINC is a program that can attach to many projects and you can just set this project to the highest priority, so when it gets work, you get work, and when it doesn't have work, you get some from elsewhere.
____________
1 Corinthians 9:16 "For though I preach the gospel, I have nothing to glory of: for necessity is laid upon me; yea, woe is unto me, if I preach not the gospel!"
Ephesians 6:18-20, please ;-)
http://tbc-pa.org

JoergF
Avatar
Send message
Joined: 20 Apr 15
Posts: 189
Credit: 224,425,186
RAC: 373,030
Level
Leu
Scientific publications
watwat
Message 45676 - Posted: 12 Dec 2016 | 16:17:39 UTC
Last modified: 12 Dec 2016 | 16:18:43 UTC

Right ... frankly I would like to help by pushing science forward rather than wasting energy. If my only interest was heating my house with GPUs I would buy a pack of GTX580 and run Furmark all day long.

Seems that the admins have underestimated the amount of work required to upgrade their server... and it doesn't make much sense to me to repeatedly comment on lacking tasks. So i have switched to Folding@Home for a while and will get back to GPUGRID in a couple of weeks.
____________
Die Liebe allein versteht das Geheimnis, andere zu beschenken und dabei selbst reich zu werden. [Clemens von Brentano]
Only love understands the secret of giving and getting richer at the same time [Clemens of Brentano]

PappaLitto
Send message
Joined: 21 Mar 16
Posts: 240
Credit: 970,319,881
RAC: 3,543,850
Level
Glu
Scientific publications
watwat
Message 45679 - Posted: 12 Dec 2016 | 18:04:51 UTC - in response to Message 45676.

So i have switched to Folding@Home for a while and will get back to GPUGRID in a couple of weeks.


Is there a BOINC app for that project, I would love to contribute to it as they support cpu and gpu but I prefer the "backup project" enviornment that BOINC offers, also I can't get Gridcoin on their app.

Wiyosaya
Send message
Joined: 22 Nov 09
Posts: 108
Credit: 143,790,703
RAC: 356,990
Level
Cys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 45681 - Posted: 12 Dec 2016 | 19:09:20 UTC

Guys, this is an R&D project. Please be patient with them. Unfortunately, they do not have unlimited professional programmers and unlimited money at their disposal. This is a big change for them, and even professional programmers may easily have underestimated the time required for this change. To make an analogy, this is akin to swapping out the engine a car for another engine.

Things will get going again.

____________

Trotador
Send message
Joined: 25 Mar 12
Posts: 82
Credit: 1,058,557,670
RAC: 977,575
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 45687 - Posted: 12 Dec 2016 | 20:54:58 UTC - in response to Message 45681.

Guys, this is an R&D project. Please be patient with them. Unfortunately, they do not have unlimited professional programmers and unlimited money at their disposal. This is a big change for them, and even professional programmers may easily have underestimated the time required for this change. To make an analogy, this is akin to swapping out the engine a car for another engine.

Things will get going again.


+1

[CSF] Thomas H.V. Dupont
Send message
Joined: 20 Jul 14
Posts: 524
Credit: 55,457,175
RAC: 59,240
Level
Thr
Scientific publications
watwatwat
Message 45699 - Posted: 13 Dec 2016 | 16:38:35 UTC - in response to Message 45687.

Guys, this is an R&D project. Please be patient with them. Unfortunately, they do not have unlimited professional programmers and unlimited money at their disposal. This is a big change for them, and even professional programmers may easily have underestimated the time required for this change. To make an analogy, this is akin to swapping out the engine a car for another engine.

Things will get going again.


+1

+2
____________
[CSF] Thomas H.V. Dupont
Founder of the team CRUNCHERS SANS FRONTIERES
www.crunchersansfrontieres.org

Erich56
Send message
Joined: 1 Jan 15
Posts: 346
Credit: 1,457,365,177
RAC: 2,705,716
Level
Met
Scientific publications
watwatwat
Message 45701 - Posted: 13 Dec 2016 | 17:18:43 UTC

Guys, what's wrong with asking "Could we have a Server and WU situation update?"

Just 1 or 2 lines once a day, and thousands of crunchers would appreciate that they get at least a minimum of information.

Profile caffeineyellow5
Avatar
Send message
Joined: 30 Jul 14
Posts: 225
Credit: 2,658,976,345
RAC: 0
Level
Phe
Scientific publications
watwatwatwatwat
Message 45717 - Posted: 14 Dec 2016 | 2:37:13 UTC - in response to Message 45676.
Last modified: 14 Dec 2016 | 2:38:11 UTC

Seems that the admins have underestimated the amount of work required to upgrade their server... and it doesn't make much sense to me to repeatedly comment on lacking tasks.
From my reading of the forums and the news section posts, I think there are 3 separate issues. I also think those 3 issues have been combined into one collective answer in the minds of the crunchers on this forum and possibly anyone who saw the announcements that went out to the clients and then saw the lack of work hit after or during around the time as the server upgrade.

1) Server upgrade: Presumed completed the day we were told it was. Apparently it is a server that has to do with storage space and not logistics, the website, or work unit creation, but only storage of finished units until they can be used for the science, summarized, and discarded. At least, this is what I understand from the information provided to us/the explanations given about it.

2) Connection issues, specifically downloads, but both up and downloads have been mentioned: Mostly the upload issues are reported around site outages only or are due to client side internet connection issues. Upload issues are not ruled out for other reasons or as a general issue with the downloads. Downloads are an issue and that is an issue that is being investigated, but apparently more in our ideas than by real people on the ground at the college. Several theories are posited by users here on the forums.

3) Amount of new work available: This goes in cycles. Sometimes we can go almost a month without any. Early and in mid 2016 and times in 2015, we had a few almost that long. Most times it is just a weekend that we can run bone dry. For the most part, we have between 500 and 4,000 live tasks in the field. When that number of live tasks combined between sent and unsent gets to about 2,000 long tasks, we hit an equilibrium of users/devices and tasks available for every request that can receive a task. Currently, we are below that equilibrium, so people think we have no tasks. This is a false perception as there are just around 1,100 of these and half that in short tasks. Maybe as much as 100 of these will fail back to the queue in between new tasks being added constantly. It does depend on when you send a request and when these tasks are in the queue as to whether you will get any of them. I have been on the good end of that as I have been getting 2-5 tasks at a time on my system that can run 6 tasks at a time. I have also been consistently getting tasks on my laptop, which takes 28-72 hours to finish a task, short or long. Some are on the bad end of that also, totally based on timing, and that is unfortunate, but fair.

Also, "to repeatedly comment on lacking tasks" is not a requirement and does not make the students make more work units in any more or less frequency than they will on their own. As a matter of fact, I am not sure if those two are inclusive to each other. They are most like still 'on their own' whether you say it or not, because I am not sure if they read these forums, each one of them. They are probably more interested in their school work, including the school work we help them do on our GPUs, than they are about reading forums that don't help reach that goal. If there is one or two students who do read them all or if the moderators on these forums only bring the major concerns to their attention, then the concerns and suggestions are getting back to them without any one or group of people repeating it anyway. The real concerns are the ones that have a few people, usually people who have 1-2 posts total, that come when something is wrong and noticed. Of the things we complain about and debate back and forth, I am sure, most of it stays between us here.

So i have switched to Folding@Home for a while and will get back to GPUGRID in a couple of weeks.
I don't know if those two things are your only options for science and your GPU's, but I have also considered it during the down times we get. I instead tend to go toward a different project for my GPUs than F@H. Either way, if you stick with something on BOINC for a science based project you believe in the work they do and are available for your GPUs, it is always recommended here on the forums that you just set your priorities for the projects so that they look for work, as independent projects, but get work and try to do this one as a priority over the others, if it gets work from here, then switches back to the other when nothing is available from here.

Guys, what's wrong with asking "Could we have a Server and WU situation update?"

Just 1 or 2 lines once a day, and thousands of crunchers would appreciate that they get at least a minimum of information.

From what I just said, there should be no need for either actually. The server could be fixed and done the day we got a message saying it was. The work units are being made and sent out and there are enough clients running that we eat them up as fast as they are spit out and are still left hungry, but that goes in cycles of school year, paper being written, our use for those papers, etc. And update past the Server Status page we already have seems redundant. We get notices when we get new students, we see their work when they send it, and we run low or out if we are in a down cycle between situations listed and probably more unlisted. That really is self explanatory when you realize the project is only here run out of a college in Spain for school work and scientific papers. The server thing is possibly explained in that it is done and installed, in which case, they told us already. And the connection issue, once they find out what is definitely wrong, it will be fixed or we will be told why it can't be. I don't think we are having any information withheld from us or delayed to us in transit somewhere. A line or two a day is a bit much to state the obvious, when you know where to look for basic information, which for these 3 issues, is these forums and the Server Status page.

As always, correct me if I am wrong on any of the assumptions I made. I get my information from these forums and make deductions about the rest from what I have available as well.
____________
1 Corinthians 9:16 "For though I preach the gospel, I have nothing to glory of: for necessity is laid upon me; yea, woe is unto me, if I preach not the gospel!"
Ephesians 6:18-20, please ;-)
http://tbc-pa.org

Betting Slip
Send message
Joined: 5 Jan 09
Posts: 574
Credit: 1,911,306,275
RAC: 1,783,457
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 45721 - Posted: 14 Dec 2016 | 10:45:26 UTC - in response to Message 45717.
Last modified: 14 Dec 2016 | 11:01:14 UTC

I asked a very straight forward question "Could we have a Server and WU situation update?"

It wasn't any type of demand or criticism.

I just feel that when any type of problem crops up we speculate endlessly, talking to ourselves.

Profile Beyond
Avatar
Send message
Joined: 23 Nov 08
Posts: 1073
Credit: 4,465,023,904
RAC: 418,532
Level
Arg
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 45726 - Posted: 14 Dec 2016 | 15:52:18 UTC - in response to Message 45721.

I asked a very straight forward question "Could we have a Server and WU situation update?"

It wasn't any type of demand or criticism.

I just feel that when any type of problem crops up we speculate endlessly, talking to ourselves.

Seems to me to be a fair question and would take less than a minute to answer. On some projects the admins are very responsive to their users (customers). On others, not so much.

Toni
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 9 Dec 08
Posts: 554
Credit: 4,273,184
RAC: 0
Level
Ala
Scientific publications
watwatwatwat
Message 45727 - Posted: 14 Dec 2016 | 16:07:30 UTC - in response to Message 45726.
Last modified: 14 Dec 2016 | 16:09:40 UTC

Caffeineyellow summarized nicely. Of the three issues:

1. Server upgrade went as planned and is completed. There are still graphical glitches on the web pages, which we hope to fix with time. Addressing these has to be balanced with scientific tasks and available hands, namely point 3 below.

2. We investigated connectivity issues with your help, and reported them with data to the network admins. There is nothing we can do on the server.

3. Lack of workunits. Of course we would like to increase the amount and diversity of experiments done. This was the main reason behind the server upgrade. Properly preparing experiments however requires time and deep consideration. We are on it. I, for one, would like to prepare further experiments, but it's not something that can be done (properly) from one day to the other.

As always, we are REALLY thankful for your support (and patience).

Profile Beyond
Avatar
Send message
Joined: 23 Nov 08
Posts: 1073
Credit: 4,465,023,904
RAC: 418,532
Level
Arg
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 45728 - Posted: 14 Dec 2016 | 16:19:23 UTC - in response to Message 45727.

Toni, thanks for the update. It's greatly appreciated!

Erich56
Send message
Joined: 1 Jan 15
Posts: 346
Credit: 1,457,365,177
RAC: 2,705,716
Level
Met
Scientific publications
watwatwat
Message 45735 - Posted: 15 Dec 2016 | 6:19:35 UTC - in response to Message 45728.
Last modified: 15 Dec 2016 | 6:20:14 UTC

Toni, thanks for the update. It's greatly appreciated!

+ 1

Post to thread

Message boards : Server and website : Could we have a Server and WU situation update?