Joined: 16 Dec 07
I'm a CPU Only Cruncher for MilkyWay and thought to contribute a few cycles while SETI@Home was still remodelling their servers.
On November 11 I noticed a servere drop in credit awarded for the number of tasks completed and reported.
None of the Tasks appeared to have errored out.
Number of Workunits I've Reported over the last few days
11-Nov-2010 16:05:01 [Milkyway@home] Reporting 4 completed tasks, requesting new tasks for CPU and GPU 12-Nov-2010 22:20:44 [Milkyway@home] Reporting 7 completed tasks, requesting new tasks for CPU and GPU 13-Nov-2010 13:01:32 [Milkyway@home] Reporting 4 completed tasks, requesting new tasks for CPU and GPU 13-Nov-2010 13:10:36 [Milkyway@home] Reporting 1 completed tasks, requesting new tasks for CPU 14-Nov-2010 15:43:23 [Milkyway@home] Reporting 5 completed tasks, not requesting new tasks
I took special note of the tasks I reported today (14-Nov-2010) and I've noticed I've been the only CPU for:
Workunit 183210055 [Task 242726316]
Workunit 183210056 [Task 242726317]
Workunit 183211484 [Task 242728396]
I also noticed that the other tasks which were validated with ATI GPUs were generated after my task had been returned.
The GPUs Validate Each Other and My CPU Task then gets discarded.
I understand the BOINC quorum validation process of generate tasks until a quorum is reached but Why is Milkyway going to satisfy a CPUs immediate request for work (within MilkyWays workunit limits) wait for it to be returned and then generate new tasks until a quorum is reached if that quorum is more than likely going to be two validating GPUs.
©2019 Astroinformatics Group