• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

This is weird

minendo

Elite Member
So I was experimenting today on my XP1800+ machine with SETI and found this to be real unusual. I decided to run two instances of SETI at the same time on my single processor and found that my work unit time increased by 20-30 minutes. I now get two work units every 3.5 to 4 hours.

What I want to know is why does running two instances at the same time complete two work units in 3.5-4 hours while running one instance takes 3-3.5 hours per work unit. Right now each process is getting around 49% of CPU usage and I am returning twice as many work units.

Anyone else notice this? If not, give it a try and see what you get. I would like to see what other people come up with.
 
Hmmm...I wonder if CPU time = REAL time to return a WU though?

If you shut down your PC at night...and it's at 50% complete with a WU....and your average is 4 hours...it will still complete the other half of the WU in 2 more hours.....and it will list CPU time as 4 hours total...even though it may have taken 20 hours (from being off) to complete....

Sometimes, I think CPU time is confusing...😕

Well, I'm always confused anyway! :Q
 
You would almost have to use a real time clock...measure how long it takes to send in a SINGLE crunching WU...and then restart (after that WU)...with two fresh WU's (two instances of the CLI)....measure how long it takes to send in those two...probably more than twice the single, as there is overhead in switching back and forth with the processor...probably why CPU time is indeed higher because of all the switching involved in a multitask situation.

Anyone else have a better explanation? 😕
 


<< Right now each process is getting around 49% of CPU usage and I am returning twice as many work units. >>


Yikes! This can't be good for me in the race! 🙁
However, it is good for SETI and TA, so if it works for you, go for it! 😀
This is what the races are all about.
 


<<

<< Right now each process is getting around 49% of CPU usage and I am returning twice as many work units. >>


Yikes! This can't be good for me in the race! 🙁
However, it is good for SETI and TA, so if it works for you, go for it! 😀
This is what the races are all about.
>>


I need to somehow keep up with the 20 WU/day you are returning.
 
We at DSLR have experimented with this a few times. See here: http://www.ameri-becca.com/cputimes2.htm

We have a couple very long threads about it too.. but unfortunatley our fourm's search engine sucks. If you have the time to look for the threads I suggest reading them, as I found them very interesting..

-Mike
 
i can just add.. that when i ran seti with 0% of cpu usage, the time on setispy did not increase at all.. even tho it was open and running.
 


<< Hmmm...I wonder if CPU time = REAL time to return a WU though?

If you shut down your PC at night...and it's at 50% complete with a WU....and your average is 4 hours...it will still complete the other half of the WU in 2 more hours.....and it will list CPU time as 4 hours total...even though it may have taken 20 hours (from being off) to complete....
:Q
>>



I'm Pretty sure CPU time is the actual tiem the CPU was working on the WU. Mine currently run about 4Hrs 5min per.

I will be very interested to find out anout the multiple processes.

What OS are you using?

Alric
 
From memory, the DSL guys said if you have a fast processor (1 Ghz +), lots of RAM (256 MB+) and a dual CPU capable OS (W2k or XP Pro) there is an advantage to running two processes simultaneously. I ran 75-100 units each way, normalized for angle ranges, and found a neglible difference. I couldn't find my post in the archices otherwise I would point you toward it.
 
Back
Top