| log in |
|
21)
Message boards :
Number crunching :
WU / task creation flow
(Message 1815)
Posted 8 May 2020 by JaviF I see. Flow of new tasks is really low. As per below Numbers, is 588 each 15 mins, around 2350 tasks / hour. Even a tiny contributor like me, who has 16 cpu cores, I require 4 tasks / hour average. 4 / 2350 = 0,17 %. Current flow could not even support 1000 contributors like me, of course as soon as someone comes with 200 cores, server wont be able to feed all users :( |
|
22)
Message boards :
Number crunching :
WU / task creation flow
(Message 1813)
Posted 8 May 2020 by JaviF Hi team, It seems from server_status, we are running out of tasks... Maybe is just a display issue, I can't say :) Thx for checking¡ Javi F |
|
23)
Message boards :
Science :
SARS-CoV-2 virus
(Message 1805)
Posted 27 Apr 2020 by JaviF 166976 today....we are almost over :) |
|
24)
Message boards :
Number crunching :
Cloud computing and hosts ids
(Message 1775)
Posted 16 Apr 2020 by JaviF Hi team, I have some cloud instances computing for TN Grid but I face huge issues when setting up a new one, as I am usually assigned an existing host id. I know this is usually because machine name and MAC are same but I only get a solution by using different OS (and not always). Is there any other way somehow, to force this? Thx¡ Javi F |
|
25)
Message boards :
Science :
SARS-CoV-2 virus
(Message 1762)
Posted 7 Apr 2020 by JaviF Im in as well. I have been monitoring the server status and it can't produce above 2K ready to send tasks. I am now attaching couple machines more (I know is not gonna be a difference but if 10 guys like me do the same, sever will be exhausted). |
|
26)
Message boards :
News :
World Gene Expansion Challenge 2020
(Message 1714)
Posted 20 Mar 2020 by JaviF It seems challenge has exhausted the server...no WU available, even though gene/day is not very high. Maybe wu creator is down? |
|
27)
Message boards :
News :
World Gene Expansion Challenge 2020
(Message 1686)
Posted 26 Feb 2020 by JaviF There won't be high participation because there will be no new user crunching "ever". We are same users always as this is listed nowhere |
|
28)
Message boards :
News :
World Gene Expansion Challenge 2020
(Message 1684)
Posted 25 Feb 2020 by JaviF I still think we should list the project in boinc. I think there is no point to keep this as a "low profile" crunching project. We all understand server has limited resources but it was added 2000 more WU to the queue as of last challenge and it is always full. By not doing this, it will take ages and ages... |
|
29)
Message boards :
Number crunching :
WU / task creation flow
(Message 1633)
Posted 4 Nov 2019 by JaviF Interesintg, less is nothing:) But I was also wondering tasks received and all that kind of "flows" in Wu/ tasks here and there. I just added cpu to perform aprox 100 tasks a day and was wondering how "much" is that :) |
|
30)
Message boards :
Number crunching :
WU / task creation flow
(Message 1630)
Posted 31 Oct 2019 by JaviF Hi team, Would it be possible to see WU creation, task sent in a day...all that kind of stats /graphs? Which would be the limit server could deal with,using current set up? 200 CPU cores more? (Example). Have a nice day. Javi |
|
31)
Message boards :
Number crunching :
Validation inconclusive.
(Message 1598)
Posted 16 Jul 2019 by JaviF No. I had Rasperri Pi for months awith 100 OK results. However, is also true, it takes around 48 hours / core to crunch a WU :) Not a problem. |
|
32)
Message boards :
Science :
Scientific results (Comments)
(Message 1580)
Posted 10 Jun 2019 by JaviF There's room enough for increasing the performance. We always have around 7K tasks ready to be sent so problem is not there but (maybe) in the reception part ( I don't think so, because capacity to send must be similar to the receiving capacity (otherwise some of the queue could grown ad infinitum ). We are less than 300 users a day sending results. It should not be a problem to double this number, starting by listing the project in Boinc by only that step, we could get lots of them, even if they crunch a dozen tasks and forget about boinc, they are valid anyway... What do you think about it? |
|
33)
Message boards :
Science :
Scientific results (Comments)
(Message 1575)
Posted 3 Jun 2019 by JaviF Great. So a total of 87554 genes /isoforms * 294 tasks / each (we dont count here resends) = 25.74 Million tasks. We need at least 2 tasks to have a "valid" result so: 25.74 * 2 = 51.48 Millions tasks. If we are doing now 26 genes a day (average) this means a returning rate of 26 * 254 ( * 2 to be valid) = 13K tasks / day? If this is the case, we definetely need to gain computing power somewhere :D |
|
34)
Message boards :
Wish List :
Tn-Grid in Boinc Manager project list
(Message 1573)
Posted 31 May 2019 by JaviF And wouldn't we be happier if we double computing capacity? With current experiment, we are talking about years... I understand there is a limit, but this is as easy to deal with by controling how many tasks we send. Now, we have always over 6K "ready" tasks... |
|
35)
Message boards :
Number crunching :
fubar host
(Message 1570)
Posted 31 May 2019 by JaviF I am under grcpool. And now number of error tasks is > 1K. Is not that difficult to notice that in the link... I wonder if server penalizes these hosts untill trust is back |
|
36)
Message boards :
Number crunching :
fubar host
(Message 1567)
Posted 30 May 2019 by JaviF Morning¡ I wonder why this is allowed... http://gene.disi.unitn.it/test/results.php?hostid=9538 |
|
37)
Message boards :
Science :
Scientific results (Comments)
(Message 1566)
Posted 29 May 2019 by JaviF By the way...It would be rather interesting to understand what is included in each "task / work unit". What is it's relation to each Gene/Isoform? We can see X tasks being crunched and Y Genes per day, number are quite different. I am just curious... Thx in advance¡ Javi |
|
38)
Message boards :
Science :
Scientific results (Comments)
(Message 1562)
Posted 23 May 2019 by JaviF It looks definetely better¡ Thx for the updates :) |
|
39)
Message boards :
Science :
Scientific results (Comments)
(Message 1560)
Posted 23 May 2019 by JaviF In that case...wouldn't be interesting to add something to "science status" link? If you go there, all is on hold and you can get no clue how far are we with current experiment (no need to know the exact % but the current information is almost useless...we dont know whether 1 day or 1 billion years is remaining). |
|
40)
Message boards :
Science :
Scientific results (Comments)
(Message 1557)
Posted 22 May 2019 by JaviF Hi, Somehow, I agree and disagree at the same time :D Technically speaking, we "know" what are we sending to crunch. There must be a "start" and "end" set of possibilities, and we also know if there is something interesting in case X, we investigate deepper (which we can't predict, I agree). Now is like crunching all genes of the world or? If we go to science status, all 3 organism are "on hold"... |