RNA World (beta) needs donations
The RNA World (beta) project needs donations to cover the BOINC project server (bandwidth and traffic) costs. Please support us and donate!
Posts by Michael H.W. Weber
|
101)
Message boards :
News :
Problems solved
(Message 175)
Posted 30 Apr 2010 by  Michael H.W. Weber
Post: The WU generator is working again, the small WUs will now be processed on the server, except for very, very few which we will deliver to some remote machines. Please also note that the server response delay has been drastically reduced such that connection issues should be resolved by now as well. We have plenty of WUs, will not run out of work and hope for your further support which is very much appreciated. ;-) Michael. |
|
102)
Message boards :
News :
WU generator transiently stopped...
(Message 174)
Posted 28 Apr 2010 by Michael H.W. Weber
Post: ...due to improvements we are currently working on to process the tiny WUs on the server in the future. The background is that batches of very short WUs cause severe server stress regarding the number of connections and MySQL requests despite our good server hardware specifications. Since these small WUs will re-occur in future projects as well, we need to do something about this. At present, we have implemented a server response delay which is causing many machines to sit idle exclusively once these small WU batches enter the system - a situation which clearly is unacceptable. These WUs are usually consisting of plasmids or phages with very short DNA sequences and are consequently processed within just a few minutes - sometimes even shorter depending on your hardware - and therefore the server contact frequencies are massively increased. We could solve the issue by just adding a couple of servers as put into practice by some of the bigger DC projects. Unfortunately, our current project budget does not allow for such a solution, so we decided to process these small WUs on one or two of the server cores in the future. Michael. |
|
103)
Message boards :
News :
Genome annotations running
(Message 172)
Posted 26 Apr 2010 by Michael H.W. Weber
Post: As you may have seen, in parallel to the tuberculosis project, we have also initiated the systematic genome annotation analyses (GA projects) in order to globally map the presence of non-coding RNAs. Michael. |
|
104)
Message boards :
News :
no multiple screensaver starts anymore
(Message 170)
Posted 21 Apr 2010 by Michael H.W. Weber
Post: Yes, the current RNA World screensaver is a Flash-driven set of videos that we have generated as a PRELIMINARY solution for those that like fancy graphics. ;-) In the FAQ we have described a work-around for the case that multiple windows are opening. If you do not like the screensaver to run (which indeed consumes up to 80% of CPU load), please simply disable it in Windows (it won't run under Linux anyway). At a future time we will seek to implement it for all operating systems in more nice way, i.e. by trying to use OpenGL and by also adding a some user information. In fact, if you know someone who is firm with programming such screensavers, just let us know. We would be very happy to have some help with this one to alleviate the current screensaver situation since at present we have to put all our programming efforts into different things more related to the scientific objectives of the RNA World project. Michael. |
|
105)
Message boards :
News :
Tuberculosis project
(Message 168)
Posted 12 Apr 2010 by Michael H.W. Weber
Post: The minimum RAM requirement for these WUs is currently set to 300 MB. For further discussions please register on our discussion forum. Michael. |
|
106)
Message boards :
News :
Tuberculosis project
(Message 166)
Posted 11 Apr 2010 by Michael H.W. Weber
Post: When will the new app show up in the preferences page? Please uncheck uppCASE (which is only a BOINC-implemented test application for project function tests) and check CMSEARCH to get the novel WUs. Michael. |
|
107)
Message boards :
News :
Tuberculosis project
(Message 164)
Posted 11 Apr 2010 by Michael H.W. Weber
Post: When will the new app show up in the preferences page? For this project we actually do not require a new application. Michael. |
|
108)
Message boards :
News :
Tuberculosis project
(Message 162)
Posted 8 Apr 2010 by Michael H.W. Weber
Post: This is just the project's NEWS section. Please, register in the discussion forum and post such issues there. Thank you very much. :) Michael. |
|
109)
Message boards :
News :
Tuberculosis project
(Message 160)
Posted 7 Apr 2010 by Michael H.W. Weber
Post: The first set of WUs dealing with our tuberculosis project (please see the previous thread and the scientific project objectives on our website) is now ready for processing. These WUs should have quite heterogeneous run times of up to a maximum of 10 hrs (on an Intel i7 920 2,66 GHz Quad-Core CPU) and should be devoid of RAM or traffic problems. I hope that even those that had problems with our last set of huge CMS WUs will consider participating again since I believe this is a very interesting research project. Michael. |
|
110)
Message boards :
News :
New work
(Message 157)
Posted 12 Mar 2010 by Michael H.W. Weber
Post: We do grant the credits wherever possible but need to make sure to treat all people the same for fairness reasons. Michael. |
|
111)
Message boards :
News :
New work
(Message 154)
Posted 9 Mar 2010 by Michael H.W. Weber
Post: This weekend .... Also, please watch the RAM requirements since a few WUs might end up taking up to 2.3 GB of memory. Thank you for the information. I will update our FAQ accordingly. Michael. |
|
112)
Message boards :
News :
New work
(Message 152)
Posted 8 Mar 2010 by Michael H.W. Weber
Post: then you should at least have granted credit for those WU's which were completed by win32 hosts. Well, there are two equally fair ways to handle this. Either we grant credit to everybody who is affected or grant credit to none. Sadly, the former is very difficult to put into practice given the development stage of the BOINC server software (how can we differentiate between WUS that crashed due to suboptimal project settings and those that crashed on the basis of other reasons?) so we decided to go for the latter hoping for some understanding by our participants who certainly put emphasis on scientific results and performance progress of this project rather than on just collection credits. The reports of the errors have finally enabled us to fix the problem such that under Linux CMC should now run error-free. I think that is valuable progress worth much more than just a few virtual points, right? Michael. |
|
113)
Message boards :
News :
New work
(Message 150)
Posted 7 Mar 2010 by Michael H.W. Weber
Post: you either accomodtate and get your project up and running by filling in the gaps or... Indeed and that is exactly what we did. Michael. |
|
114)
Message boards :
News :
New work
(Message 148)
Posted 7 Mar 2010 by Michael H.W. Weber
Post: hmm - something did work. i got 5 of them on a host running srv2k3x32.. Well, it is not really encouraging, I know. But I guess you understand that this complaint would have to be addressed to the BOINC development team rather than to us since at least to our understanding it is caused by a yet unknown bug in the BOINC server software. The cancellation is exactly the mandatory measure to avoid further usefless energy consumption and now it all seems to run smoothly. Michael. |
|
115)
Message boards :
News :
New work
(Message 146)
Posted 7 Mar 2010 by Michael H.W. Weber
Post: hmm - something did work. i got 5 of them on a host running srv2k3x32.. That should not have happened. It seems there is some bug with the BOINC server software which we are going to hunt now. Thank you very much for informing us immediately. Please delete all CMCs on x86 machines wherever possible. Michael. |
|
116)
Message boards :
News :
New work
(Message 144)
Posted 6 Mar 2010 by Michael H.W. Weber
Post: This weekend we will release new work. As announced earlier, first we will send out two sets of CMC WUs exclusively for Windows x64 machines. These will be submitted this evening. In parallel to that we will upload our first CMS test set of WUs to analyze eukaryotic organisms including the human genome to all systems. Please note that these WUs will require a broadband internet connection since the sizes of the genomes of these organisms are one to two orders of magnitude larger than all what we have analyzed so far. Importanly, run times can be significantly longer as well with up to 100 hrs for a single WU. Also, please watch the RAM requirements since a few WUs might end up taking up to 2.3 GB of memory. Michael. |
|
117)
Message boards :
News :
Problem solved
(Message 143)
Posted 5 Mar 2010 by Michael H.W. Weber
Post: I have good news for those of you whose machines failed on the CMC "monster WUs" delivered to Linux x64 systems. Tonight we most likely solved the problem which indeed has to do with memory allocation. A new parameter setting has been applied and these WUs are now out again in the wild for those who dare munching on them once again. Solving this problem was only possible with the valuable help of some of you who ran the diagnostics which we had posted to diverse forums. Thanks again for your help. Michael. |
|
118)
Message boards :
News :
Current status
(Message 140)
Posted 4 Mar 2010 by Michael H.W. Weber
Post: The entire set of CMC work units for Linux x64 machines is almost complete now and so is the current CMS bundle. We have already prepared the first eukaryotic organisms which will cause significantly more traffic during download and correspondingly require more HD storage as well. As soon as the results have been assembled (a matter of just a few days), we will normalize the client server communication intervalls such that all the problems that occurred with the last CMS work units will not reappear. We will then release a set of CMC work units for Windows x64 only plus a first test set of the eukaryotic CMS WUs for all systems to check whether it runs smoothly. If that is the case, we will fill up the server with more work. Exact dates will be announced here. Thank you all for your patience and help with the trouble shooting. Michael. |
|
119)
Message boards :
News :
Run-time filter required
(Message 139)
Posted 25 Feb 2010 by Michael H.W. Weber
Post: We actually know the sources of the issue and have solved it. Michael. |
|
120)
Message boards :
News :
Run-time filter required
(Message 137)
Posted 24 Feb 2010 by Michael H.W. Weber
Post: First of all, our traffic is always in compressed format to reduce bandwidth. Secondly, as said before, we have thought about combinatorial WU archives. It is not advisable because it would require to significantly change the validator. That in turn makes not much sense as we will not send out archives with so many WUs that are this short again. We have learnt a "MySQL lesson" from this and that's what our current testing phase is meant for - although we are processing scientifically meaningful work throughout our "test" runs. ;-) Michael. |
Previous 20 · Next 20