| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
| World Community Grid Forums
|
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 4
|
|
| Author |
|
|
RicktheBrick
Senior Cruncher Joined: Sep 23, 2005 Post Count: 206 Status: Offline Project Badges:
|
I have been watching the results returned from yesterday. It use to be that IBM did more results than the next 30 or so members. But today they did not even finish first as it seems that everyone in the top 10 had a huge increase in their output except for IBM. There was an increase of over a half of a million results today over yesterday and it seems that all of that increase came from the top 10 members as they were responsible for over a third of all the results(over 725,000 results). If just 20 more members could duplicate the top 10 there would be no need for the rest of us as I think they could handle all the work units that WCG could generate. Marist college had an increase of around 45,000 a day to 275,000 a day. Maybe they just got a super computer.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I'd wait a few days and then take an average. Some recent OET batches have been completing in only a few minutes. Just when those show up in the results depends on a cruncher's queue length.
|
||
|
|
marist_university
Advanced Cruncher USA Joined: Mar 30, 2005 Post Count: 107 Status: Offline Project Badges:
|
We're focusing on OET right now, which has caused significant bump in results returned. Before this new wave of OET, we were at about 50k/day, then when we started focusing on it, we jumped to 100k+/day. Yesterday was an anomaly at 275k.
----------------------------------------No changes in our landscape really...and no new supercomputer ;) We do have a few mainframes, but I'm not aware of any z/VM BOINC client to run on them. Our contributions are from our desktops, laptops, and tower workstations. ![]() |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
There's a ready VBox client package for Windows [if that's what your mainframes run]. This creates a Linux environment to be able to run projects such as LHC/Atlas: Get it at http://boinc.berkeley.edu/download_all.php . If what you run is z/VM Hypervisor, here some discussion.
Read of some members running multiple VM BOINC on a single machine, supposing they then run their BOINC as if Linux. The gain is for them that even if having a Windows machine. They can crunch FAHV and OET1 on Linux, where these run much much faster... 60+% on a pure Linux device. The whole explosion of results was driven by the short OET1 and FAHV too. Latter is said to be sized up in the near future. As FAAH Autodock dropped off which ran much longer, being replaced by FAHV the overall impact was big. It's though temp. Yesterday mean OET times were 0.7 hours, this morning already up to 1.1 hrs/result. |
||
|
|
|