Posts by Abel
1)
Message boards : Number crunching : Test WUs ( Message 4029 )Posted 3350 days ago by Abel
Hmmm, we do have an intel mac. Will give it a look. |
2)
Message boards : Number crunching : Test WUs ( Message 4027 )Posted 3350 days ago by Abel Your new '1abe' work units all fail with segmentation errors on my Linux AMD computers. I noticed you have two computers with identical architectures yet one with a different linux kernel (one of which is the offending seg faulter). Can you confirm that both are crashing with 1abe? It looks like only one has gotten the 1abe complex, we will send some more WU's over next week and if you could check if your other AMD gets validated that would be awesome. Thanks! |
3)
Message boards : Number crunching : Test WUs ( Message 4026 )Posted 3350 days ago by Abel I have one Windows box that cannot receive work. It gives the following message: Nothing wrong with your cpu, we indeed support this architecture. We will address this on Monday. Thanks! |
4)
Message boards : Number crunching : Test WUs ( Message 4025 )Posted 3350 days ago by Abel I just got a bunch on the Mac and they still all fail instantly. I've seen that some others recently crunched successfully on their Macs and would appreciate any ideas. Stack trace etc. is reported on the failing tasks. This is the host's tasks page. Looks like the Xeon's are being put together with the intel core 2's. There is obviously a divergence in the two. We will reevaluate the Xeons HR. |
5)
Message boards : Number crunching : Test WUs ( Message 4024 )Posted 3350 days ago by Abel I have one Windows box that cannot receive work. It gives the following message:My G5 Mac got the same thing last night, and then was apparently told to go away for 24 hours. (I’m guessing as to the interval: I recently installed BOINC v5.10.45 and it doesn’t report communication deferrals the way my previous version did.) As far as Power PC architecture goes, we will no longer be supporting that architecture. We don't have a power pc in our lab and haven't been able to get one. It's getting harder and harder to find those machines. Anyone have any suggestions as far getting a power pc binary? As far as the windows machine not getting work, there is definitely something wrong with that. We will look into all these things on Monday. Don't forget to vote on Monday. |
6)
Message boards : Number crunching : Cobblestones ( Message 4011 )Posted 3351 days ago by Abel Obviously there are different ways to grant credit to you guys. It has been long debated on how that should be done. One method takes into account time, it you take more time on a workunit, you get more credit. This however gives advantages to slow machines. They are doing less work yet getting more credit. The other approach is FLOPS. you are granted credit based on total FLOPS of an application. This would favor fast machines over slow ones, as they have a higher throughput. There is also a hybrid of the two, basing it on the speed of your machine, the total FLOPS and the time taken. Personally I think throughput is the bottom line. And credit should be assigned by the size of the workunit in FLOPS. But in the spirit of democracy we will elect a method to compute points. Here is how it will go. In the first round every member is entitled to propose how the credits will be granted. We will then take a vote on the one you guys like the best. We well then use that scheme in assigning credits. Remember that the proposals should be for what to base the credit on and not how much credit to give. For example: Credits = total flops / flop rate is ok. But Credits = 1 gazzilion per work unit is not ok. Hope this works out, have fun. Proposals will be open until Sunday night, we will vote on Monday. |
7)
Message boards : Number crunching : Test WUs ( Message 4009 )Posted 3351 days ago by Abel Btw: I've tried Docking on my old (and lame *grin*) PIII 450 Mhz, running under Win2K. Has to be a divergency issue. May need to put your PIII in another HR class. We will look into it. |
8)
Message boards : Number crunching : Test WUs ( Message 4008 )Posted 3351 days ago by Abel On my Host 1526 I am showing seven work units, numbers 1887 , 1904 , 1913 , 1920 , 1924 , 1935 , 1940 . I do not have these work units, they downloaded to my computer (well I could locate 4 of them in the logs), and all errored out with the error code 193 (as per my message logs), but none of this information has uploaded to your servers, and you still show them as being 'Pending'. Thanks for the updates on the buggy WUs. As far as the points go, somehow they are not being normalized, we will take a look at this ASAP. |
9)
Message boards : Number crunching : Screensaver ( Message 3997 )Posted 3352 days ago by Abel Screen saver seems to have been a success! Robert did a very good job putting it together. Thank you all for your help. We hope to have it working on Linux and Mac very soon. |
10)
Message boards : Number crunching : Test WUs ( Message 3996 )Posted 3352 days ago by Abel Looks like we still have some issues to iron out. We are planning to update the version of CHARMM so we are not too worried about the seg fault issue since we already address some of those issues in the update. We just wanted to make sure HR and the screensaver where working. It seems the screensaver is fine but HR is still being difficult. We are working overtime to get it back in business. Thanks for the updates. |
Next 10 posts