| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: FightAIDS@Home
Thread: FightAIDS@Home workunit runtime increased by 20% |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 18
|
|
| Author |
|
|
armstrdj
Former World Community Grid Tech Joined: Oct 21, 2004 Post Count: 695 Status: Offline Project Badges: 
|
We have increased the runtime for all FightAIDS@Home workunits by about 20%.
Thanks, armstrdj |
||
|
|
Falconet
Master Cruncher Portugal Joined: Mar 9, 2009 Post Count: 3297 Status: Offline Project Badges: 
|
Why all these runtime increases?
----------------------------------------Is more work being put into workunits? AMD Ryzen 5 1600AF 6C/12T 3.2 GHz - 85W AMD Ryzen 5 2500U 4C/8T 2.0 GHz - 28W AMD Ryzen 7 7730U 8C/16T 3.0 GHz |
||
|
|
pramo
Veteran Cruncher USA Joined: Dec 14, 2005 Post Count: 711 Status: Offline Project Badges: 
|
My WAG is that more work is put in, possibly a bit less workload on the servers as far as number of connections are concerned?
---------------------------------------- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Think this the another IBM PTF... indeed by lengthening the jobs where the Techs can, the number of uploads queued reduce and thus if these overloads happen and recover from it, there are much less trying to push up the result files. The one that was really radical is GFAM at ~90% extension when there are presently just 20,000-22,000 results a day... run time from averaging 3.8 hours to 7 hours. Maybe it's a hint that soon the scientist receiver side issue is resolved and the supply/priority can be returned to normal level.
BTW Like with FAAH, these longer jobs had been in the pipe for a little. The average times for GFAM had already climbed from 3.8 to 5.0 in the past 7 days, not that a member looks at yesterdays' stats and does an OMD, fearing there will be 9 hour averages coming. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Think this the another IBM PTF... indeed by lengthening the jobs where the Techs can, the number of uploads queued reduce and thus if these overloads happen and recover from it, there are much less trying to push up the result files. The one that was really radical is GFAM at ~90% extension when there are presently just 20,000-22,000 results a day... run time from averaging 3.8 hours to 7 hours. Maybe it's a hint that soon the scientist receiver side issue is resolved and the supply/priority can be returned to normal level. But is that the true motivation behind the changes? ; |
||
|
|
mikey
Veteran Cruncher Joined: May 10, 2009 Post Count: 824 Status: Offline Project Badges:
|
Think this the another IBM PTF... indeed by lengthening the jobs where the Techs can, the number of uploads queued reduce and thus if these overloads happen and recover from it, there are much less trying to push up the result files. The one that was really radical is GFAM at ~90% extension when there are presently just 20,000-22,000 results a day... run time from averaging 3.8 hours to 7 hours. Maybe it's a hint that soon the scientist receiver side issue is resolved and the supply/priority can be returned to normal level. But is that the true motivation behind the changes? ; Is this a prelude to more stats badges too?   |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Think this the another IBM PTF... indeed by lengthening the jobs where the Techs can, the number of uploads queued reduce and thus if these overloads happen and recover from it, there are much less trying to push up the result files. The one that was really radical is GFAM at ~90% extension when there are presently just 20,000-22,000 results a day... run time from averaging 3.8 hours to 7 hours. Maybe it's a hint that soon the scientist receiver side issue is resolved and the supply/priority can be returned to normal level. But is that the true motivation behind the changes? ; Sigh. *All* past extensions and shortenings have been related to load control. Do you want the techs to spell it out again for you? |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
But is that the true motivation behind the changes? Sigh. *All* past extensions and shortenings have been related to load control. Do you want the techs to spell it out again for you?; There are many reasons, rationale, and motivation behind a given action or event. Which one of them applies? All of them? One or two of them? All except one or two? The obvious one? The ambiguously discussed earlier one? The not-so-much-discussed-but-potentially-important one? The apparent one that got the spotlight at the forum the past few weeks? Last month's standing issue? Last year's speculation? Which one? How do you then resolve the ambiguity? My answer: State it. WCG's answer: Remain silent. ; |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
But is that the true motivation behind the changes? Sigh. *All* past extensions and shortenings have been related to load control. Do you want the techs to spell it out again for you?; There are many reasons, rationale, and motivation behind a given action or event. Which one of them applies? All of them? One or two of them? All except one or two? The obvious one? The ambiguously discussed earlier one? The not-so-much-discussed-but-potentially-important one? The apparent one that got the spotlight at the forum the past few weeks? Last month's standing issue? Last year's speculation? Which one? How do you then resolve the ambiguity? My answer: State it. WCG's answer: Remain silent. ; This was a concerted 6 science extend action, and the obvious has been stated [by me] for the benefit of those who came to look in. Nothing needs further adding. |
||
|
|
Steve W
Advanced Cruncher Joined: Dec 9, 2005 Post Count: 110 Status: Offline Project Badges:
|
I just hope that the slower clients will still be able to report back before the report deadline or are these deadlines also going to be extended by the appropriate amount to allow time for crunching the larger WUs?
 |
||
|
|
|