climateprediction.net home page
Credit Status

Credit Status

Message boards : Number crunching : Credit Status
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 · 6 . . . 10 · Next

AuthorMessage
MartinNZ

Send message
Joined: 22 Mar 06
Posts: 144
Credit: 24,695,428
RAC: 0
Message 56909 - Posted: 21 Sep 2017, 21:45:09 UTC - in response to Message 56898.  

.... but as some people here have pointed out, it's not mission-critical.


Hi Graham, that seems to be a fairly narrow view of mission-critical and I'm not sure it is quite what WB8ILI was saying when looking at the project as a whole. When viewed over the geological timescale, our current movement in climate change is indeed mission critical. No, we are not going to have sea level flooding major cities within my lifetime, but unless we make significant movement within our generation, many thousands of people will be flooded out of their homes over the coming decades.

Of course the biggest hindrance to climate action is political inaction and if we could use distributed computing to alter the mind-state of the deniers, then I would switch instantly. But we can't and the best we can do with current resources is to help get the science and future predictions to be more accurate.

Having just looked through the list of BOINC projects, I see the CPDN being more mission-critical than any of the others, by far. If you believe in climate change and want to dedicate CPU cycles there is in my view no more deserving cause.

OK the credit issue needs to get sorted, but I don't see it as something that should stop us crunching numbers.
ID: 56909 · Report as offensive     Reply Quote
WB8ILI

Send message
Joined: 1 Sep 04
Posts: 161
Credit: 81,421,805
RAC: 1,225
Message 56913 - Posted: 22 Sep 2017, 11:47:31 UTC

Maybe "mission critical" was a poor choice for a term. What I tried to say is that one can judge how "important" some problem is by the urgency to get it fixed.

In my previous example, anything that is broke on Friday afternoon but can wait until the following week is probably not that important. If people work on it until it is fixed, even if it is "the weekend", the problem would be perceived as important.

The credit thing might not be that important because I assume there are more important items to be addressed although I don't know that for sure.

It would seem that there is some "Project Manager" or "Project Administrator" type that would have as part of their job description to be the "PR Front" for the project. That person seems to be missing in action.
ID: 56913 · Report as offensive     Reply Quote
ian

Send message
Joined: 23 Feb 06
Posts: 20
Credit: 6,743,965
RAC: 7,972
Message 56914 - Posted: 22 Sep 2017, 15:59:05 UTC - in response to Message 56872.  

You have totally missed the point - for ANY important experiment PR is vital. It is decidedly lacking here. Moreover there seems to be a feeling in this project ( not the others to which I contribute) from the posts that I read that " yes it is all right they are doing good science so come what may". Well, if you are an ardent "tree hugger", so be it. BUT we all,I hope, believe that this set of climate modelling IS taking humanity forward. So this project should make some effort to persuade not just the peer reviewers of their academic papers but to the rest of us that THIS IS THE CASE. That starts by treating its contributors with some RESPECT. Secondly by keeping them properly informed. Thirdly by spending time ( and yes money!!) in ENCOURAGING people to volunteer ( and from the look of the stats on the website there seem to be a large number who have ceased doing so).

Apologies about the diatribe BUT I do strongly support the basis of the science, I just think THAT COULD BE HANDLED A GREAT DEAL BETTER as regards the relations with stakeholders and as I said in my previous post I DO NOT think this is enhancing the reputation of the university.
ID: 56914 · Report as offensive     Reply Quote
Profile Iain Inglis
Volunteer moderator

Send message
Joined: 16 Jan 10
Posts: 1079
Credit: 6,904,049
RAC: 6,657
Message 56916 - Posted: 22 Sep 2017, 18:01:14 UTC

Oddly enough it was exactly two years ago today that the end of Hannah Rowlands position as project communications officer was announced (here). That appointment worked well and I assume the lack of a successor is down to money; perhaps it's hard to write a science budget proposal that includes PR, such is the unusual nature of a distributed computing project - I don't know. The project front page has improved in recent years, with more information about projects and various topical feeds (Twitter and job batches). But although credits are sometimes treated as a proxy for project commitment to its volunteers, the PR aspect is really a separate matter.

The current credit hiatus is a bit like the recent global warming hiatus: just as the heat is not lost so the trickles on which credits are based are not lost - the problem is visibility. In other words the trade-off is not science vs credits, it's science vs credits-now, because they will turn up at some point. That's the bit I don't get: credits are part of the BOINC deal and should be granted, but as long as the issue is being addressed I don't really understand why it matters when.

The project staff know about the problem and they're working out a fix. This time it isn't a question of just giving a machine a kick - it's more difficult than that. Unfortunately, we'll just have to wait.
ID: 56916 · Report as offensive     Reply Quote
CJ Xuereb

Send message
Joined: 24 Oct 16
Posts: 6
Credit: 1,735,498
RAC: 853
Message 56926 - Posted: 23 Sep 2017, 1:30:29 UTC - in response to Message 56898.  
Last modified: 23 Sep 2017, 1:39:33 UTC

Having read through this thread, my conclusion is that the people running the project don't care enough about us number crunchers to communicate directly with us. All the information we get - whether about credit or problems with the project - seems to come from volunteers/moderators (for which many thanks).


Totally agree with Grahamt.

What may be even more disturbing is that they may not have the people with the skills to solve this problem, and that months may go by before they finally decide to bring in someone who can fix the problem.

Even worse than the above, is that all the CPU cycles I am currently donating to this project will be wasted if this problem is never solved and the project goes under.

I plan to do the same as Grahamt. Let my current CPDN units complete and hopefully, by then, WCG will have started their own Climate Change Research projects.

http://www.worldcommunitygrid.org/about_us/viewNewsArticle.do?articleId=529
ID: 56926 · Report as offensive     Reply Quote
Darmok

Send message
Joined: 29 Dec 09
Posts: 34
Credit: 18,395,130
RAC: 0
Message 56935 - Posted: 24 Sep 2017, 3:24:08 UTC - in response to Message 56916.  

It is now getting to two months and they are obviously not solving their problem.

The basis of BOINC is to "reward" the contributors. Failing this, I would suggest perhaps it is time for people overlooking BOINC to assist and/or do something about this, and then informing us, not only through this forum, but perhaps via direct email to all contributors with outstanding credit. Now that would be needed good PR.
ID: 56935 · Report as offensive     Reply Quote
Art Masson
Avatar

Send message
Joined: 16 Oct 11
Posts: 254
Credit: 15,954,577
RAC: 19
Message 56945 - Posted: 24 Sep 2017, 15:10:23 UTC

I am a long term contributor to CPDN, however I am now planning to stop processing CPDN Work Units starting October 1, 2017 until a credit run occurs. Thanks in advance.
Art Masson
ID: 56945 · Report as offensive     Reply Quote
Nigel Cade

Send message
Joined: 30 Apr 07
Posts: 2
Credit: 4,289,327
RAC: 3,105
Message 56947 - Posted: 24 Sep 2017, 15:47:33 UTC - in response to Message 56935.  

I don't understand this credit "thing"; it's not as if they have a tangible value, like Nectar points or airmiles. You can do nothing with them apart from "my total is bigger than yours" and then what.
The science is still being done and still needs to be done.
ID: 56947 · Report as offensive     Reply Quote
Profile fasanzaf

Send message
Joined: 4 Sep 04
Posts: 12
Credit: 7,917,448
RAC: 2,085
Message 56952 - Posted: 24 Sep 2017, 17:50:25 UTC
Last modified: 24 Sep 2017, 18:04:31 UTC

Nigel, the issue of credit is not about competition but recognition of our contribution. Is that so bad?

On the other hand, the message that the important thing is science and not the numbers begins to bore, don't you realize it? Are they asking us for blind faith? blind support? Are we talking about science or about religion or politics? It is a very pragmatic matter.

Finally, what would happen if the contribution of the volunteers goes down so much that it becomes mission-critical? How much time will they (scientists) need to wait to take action?

Please check you all this project status info page, sort it by clicking twice in Project status column header to see Production above, and tell me what do you see... (btw that stuff about BOINC Validation Pending has nothing to do with credit status updates, as stated by someones in another recent thread, is hard to believe: in my BOINC CPDN page, when my credits are updated, my validation pending becomes 0)
ID: 56952 · Report as offensive     Reply Quote
Profile Iain Inglis
Volunteer moderator

Send message
Joined: 16 Jan 10
Posts: 1079
Credit: 6,904,049
RAC: 6,657
Message 56953 - Posted: 24 Sep 2017, 18:13:20 UTC

The irony of this situation is that the project is in its current technical bind precisely because it tried to invent a method of allocating credits that respected volunteer efforts (i.e. trickles) rather than volunteer results (i.e. completed models). That system has become so computationally onerous that credits can be allocated only once a week - and now some database limit has been breached. Clearly something has to give.
ID: 56953 · Report as offensive     Reply Quote
Profile Dave Jackson
Volunteer moderator

Send message
Joined: 15 May 09
Posts: 4314
Credit: 16,378,503
RAC: 3,632
Message 56954 - Posted: 24 Sep 2017, 19:54:39 UTC

Validation Pending has nothing to do with credit status updates, as stated by someones in another recent thread, is hard to believe: in my BOINC CPDN page, when my credits are updated, my validation pending becomes 0)


As one of the people who have posted the lack of relationship, I would guess that the relationship between the two events is that the credit script sets the validation pending to 0 when it runs so that it only needs to add new results next time it runs.
ID: 56954 · Report as offensive     Reply Quote
Nigel Cade

Send message
Joined: 30 Apr 07
Posts: 2
Credit: 4,289,327
RAC: 3,105
Message 56958 - Posted: 25 Sep 2017, 10:28:31 UTC - in response to Message 56952.  

No still don't get it. It's not as if I'm sat with paper and pencil doing the maths. The cpu, on my motherboard, is non sentient and does not get bored; it just moves electrons around, or that is my belief. When it comes down to it an awful lot of science comes down to blind faith.
Perhaps the answer is to strip the credit out altogether and leave people to feel good or not about their contribution.
Thanks for the tutorial on how to use the website.
ID: 56958 · Report as offensive     Reply Quote
Jean-David Beyer

Send message
Joined: 5 Aug 04
Posts: 1045
Credit: 16,506,818
RAC: 5,001
Message 56964 - Posted: 25 Sep 2017, 19:52:26 UTC

If the problem is the complexity and computational demands in posting trickles, why not just eliminate the scoring of trickles. Let trickles be uploaded if that is useful to the scientists, but do not use them for the rankings. Just do the total results.
ID: 56964 · Report as offensive     Reply Quote
Profile Iain Inglis
Volunteer moderator

Send message
Joined: 16 Jan 10
Posts: 1079
Credit: 6,904,049
RAC: 6,657
Message 56965 - Posted: 25 Sep 2017, 20:05:27 UTC - in response to Message 56964.  

If the problem is the complexity and computational demands in posting trickles, why not just eliminate the scoring of trickles. Let trickles be uploaded if that is useful to the scientists, but do not use them for the rankings. Just do the total results.

... the trickle method was established when CPDN runs were very long indeed, and if a failure occurred before the end it was a bit unfair to grant no credits at all. Work units on this project are still quite long and the same logic might still apply,

As it happens a bug in the Mac version (now fixed) disabled trickle generation for one of the application versions and Mac users were effectively running on the kind of system you suggest - all credits for a model or no credits. However, Mac users knew that a fix was on the way.
ID: 56965 · Report as offensive     Reply Quote
_Ryle_

Send message
Joined: 17 Aug 05
Posts: 21
Credit: 15,091,561
RAC: 22,127
Message 56966 - Posted: 26 Sep 2017, 0:00:47 UTC - in response to Message 56947.  

I don't understand this credit "thing"; it's not as if they have a tangible value, like Nectar points or airmiles. You can do nothing with them apart from "my total is bigger than yours" and then what.


This logic isn't true anymore, since it is now possible to earn Gridcoin while running Boinc-projects. This project isn't whitelisted though, most likely due to these issues it has with the credits not being distributed regularly.

This project may be one of the most important ones in Boinc, but sadly it hasn't run so well in a long time. Lately those linux tasks crashing the Boinc-client altogether, but that is another topic.
ID: 56966 · Report as offensive     Reply Quote
crashtech

Send message
Joined: 1 Jun 17
Posts: 13
Credit: 23,919,122
RAC: 1,141,751
Message 56974 - Posted: 26 Sep 2017, 13:35:51 UTC

The credit system can be valuable when fine-tuning resource allocation. Since I'm juggling 26 different BOINC projects on 11 different machines, the stats generated by the points system are a valuable tool. I'm going to be ramping down my contribution to Climate Prediction for a while, until this issue is resolved.
ID: 56974 · Report as offensive     Reply Quote
Art Masson
Avatar

Send message
Joined: 16 Oct 11
Posts: 254
Credit: 15,954,577
RAC: 19
Message 56975 - Posted: 26 Sep 2017, 16:02:25 UTC

I also depend on the credit system to allocate and manage my 6 computers' utilization across 5 different projects. This is why I will stop CPDN processing on October 1st unless the ongoing problems are resolved by then.

Hopefully the CPDN technical folks will resolve the problem shortly (even if that means simplifying the credit calculations). I wish the project team the best, but the continuing lack of communication other than through the volunteer moderators is definitely disheartening.

Sincerely,

Art Masson
ID: 56975 · Report as offensive     Reply Quote
w1hue

Send message
Joined: 31 Aug 05
Posts: 20
Credit: 1,969,695
RAC: 0
Message 57004 - Posted: 1 Oct 2017, 4:16:10 UTC - in response to Message 56872.  


You have to understand the nature of this work. This is not a "mission critical" computer endeavor. This not like credit card processing, or bank account processing, or medical record processing. If things are "messed up", it can wait until whatever.

In that case, they can wait until "whatever" for me to again donate my computer time to their "unimportant" project.
ID: 57004 · Report as offensive     Reply Quote
ian

Send message
Joined: 23 Feb 06
Posts: 20
Credit: 6,743,965
RAC: 7,972
Message 57015 - Posted: 2 Oct 2017, 13:21:37 UTC

So after all the dialogue on this and other threads over the last month or so nothing positive has happened. It occurs to me that, from a number crunchers point of view, the problem we have is that this project has, to date, had too many of us to deal with the work flow. How many times have we sat with no work forthcoming? Perhaps previous comments here are right - create a scarcity of computing power and then someone might do something. I am severely reducing (but not to zero because i want to know if this thing is still running) the resource allocated by me to this project basically as a protest. I suggest others do likewise.
ID: 57015 · Report as offensive     Reply Quote
Profile Dave Jackson
Volunteer moderator

Send message
Joined: 15 May 09
Posts: 4314
Credit: 16,378,503
RAC: 3,632
Message 57016 - Posted: 2 Oct 2017, 13:49:54 UTC - in response to Message 57015.  

the problem we have is that this project has, to date, had too many of us to deal with the work flow. How many times have we sat with no work forthcoming?


While in the past there have been more cores available to crunch than tasks available, most of the time recently that has not been the case. However no withdrawal of computers is going to change the erratic nature of the availability of work from this site as that is down to the researchers around the world submitting work to Oxford for distribution. This is often though not always down to deadlines of a PhD thesis etc. These researchers or probably more accurately their universities pay Oxford University for getting the work out to us crunchers and returning the results. If they were to pay more for the services they get perhaps the credit problem could be sorted out more quickly but I suspect that so long as they get the results they need, sorting that out will not be high on their list of things to pay for. This means that work on the credit system has to be fitted in when there is space from managing the batches of tasks sent to Oxford.

Perhaps what is needed is one of the periodic gaps in work to sort it out?
ID: 57016 · Report as offensive     Reply Quote
Previous · 1 · 2 · 3 · 4 · 5 · 6 . . . 10 · Next

Message boards : Number crunching : Credit Status

©2024 climateprediction.net