| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Community
Forum: Chat Room
Thread: Question about task estimates |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 5
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi.
I'm a novise here, and have just a few questions. When I started the estimates were pretty right. but suddenly a few days ago, the estimates went up pretty much. example: Computing for clean water uses normaly about 2 hours of elapsed time, while the estimate says 32 hours before it starts. Go fight against malaria estimates 152 hours while statistics says: I have delivered 71 results, using 0:024:06:24:25 which would give an average of 8.2 hours of executing for each results. The cpu benchmark says: Number of CPU's 8 2651 floating point MIPS (whetstone) per CPU 7289 integer MIPS (Drystone) per CPU I use a SSD which has 10GB allocated for crunching. Another question: In the status field some of the tastks says: Running, high priority What does that mean? |
|||
|
||||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello Svein52,
The time estimates are made for a 'standard' computer by each project. They sometimes are far off but this seldom happens for our current projects. However, each of your computers has a local correction factor which it uses to convert the 'standard' computer time estimate into a local time estimate. This is where things normally go wrong. I occasionally find my local estimates doubling, but they quickly correct back to normal. Sounds as though your computer really glitched up the time correction factor. Just leave it alone and it will correct itself. Running high-priority means that BOINC has noticed that the time estimates suggest that it may not be able to return some results before their deadline. The BOINC scheduler algorithm on your computer switches to a different set of rules. Since you will be able to return your results on time, this is just an interesting variation. Lawrence |
||
|
|
Ingleside
Veteran Cruncher Norway Joined: Nov 19, 2005 Post Count: 974 Status: Offline Project Badges: 
|
The time estimates are made for a 'standard' computer by each project. They sometimes are far off but this seldom happens for our current projects. However, each of your computers has a local correction factor which it uses to convert the 'standard' computer time estimate into a local time estimate. This is where things normally go wrong. There's now also server-side per-application per-computer estimates, these kicks-in then at least 10 validated results for application/computer-combination, and it can look like these in some instances gives too high estimates before they're becoming more correct.  "I make so many mistakes. But then just think of all the mistakes I don't make, although I might." |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Could it be because I added a snail computer to the same user -id?
The number2 computer is running with 2 cpu's. 2494 floating point MIPS(Whetstone) per CPU 4447 integer Mips (Drystone) per CPU Is this having an effect also for the no1 computer I use? No1 produces 17000-26000 points per day while no2 produces from 1000 to 2900 points per day. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello Svein52,
Your new computer has nothing to do with it. As Ingleside notes, the March upgrade to the server code has changed a lot of things and has received a lot of bug fixes. So all this is in flux. Wait a week or so and see how things settle down. Lawrence |
||
|
|
|