Index  | Recent Threads  | Unanswered Threads  | Who's Active  | Guidelines  | Search
 

Quick Go »
No member browsing this thread
Thread Status: Active
Total posts in this thread: 19
Posts: 19   Pages: 2   [ 1 2 | Next Page ]
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 5037 times and has 18 replies Next Thread
Randzo
Senior Cruncher
Slovakia
Joined: Jan 10, 2008
Post Count: 339
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Huge difference between Elapsed time and CPU time

Hello,

I would like to ask you if you observe such huge difference between Elapsed time and CPU time on your devices as well.

I have Samsung Galaxy S III, fully updated. BOINC is set to use 4 cores, compute on battery and no Memory limit (I have always over 100 MB free), I do not use any resource intensive apps except of BOINC of course an I use the device very rarely.

The run times are in bold.

FAHV_ x3NF6_ A_ IN_ LEDGF_ 0049764_ 1302_ 2-- localhost Pending Verification 7/31/13 05:17:31 8/1/13 05:11:57 1.56 / 11.29 35.7 / 0.0
FAHV_ x4I7G_ RT_ NNRTIadj_ wNNRTI_ 0046102_ 0359_ 2-- localhost Pending Verification 7/31/13 05:17:04 8/1/13 05:11:57 2.28 / 10.72 33.9 / 0.0
FAHV_ x3ZT3_ A_ IN_ LEDGF_ 0046992_ 1038_ 1-- localhost Pending Validation 7/29/13 17:48:31 7/31/13 05:10:22 1.52 / 8.34 28.7 / 0.0

Do you experience such difference as well?

What might be the explanation?

I read an article that Android is not able to effectively execute more threads simultaneously (but the article was by Intel and in that time there were no devices with Intel CPUs running Android). Hopefully this is not true.
[Aug 4, 2013 5:57:42 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Sgt.Joe
Ace Cruncher
USA
Joined: Jul 4, 2006
Post Count: 7846
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Huge difference between Elapsed time and CPU time

If you are computing on battery power, there may be a setting (power saver) which curtails your processing when battery reserves drop below a predetermined level. Does this also occur when if the device is always plugged in ?
Cheers
----------------------------------------
Sgt. Joe
*Minnesota Crunchers*
[Aug 4, 2013 8:03:36 PM]   Link   Report threatening or abusive post: please login first  Go to top 
dskagcommunity
Senior Cruncher
Austria
Joined: May 10, 2011
Post Count: 219
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Huge difference between Elapsed time and CPU time

I have the same. same handy, pluged into power, 100% cpu.
----------------------------------------
http://www.research.dskag.at
Crunching for my Dog who had "good" Braincancer.


[Aug 4, 2013 8:16:50 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Huge difference between Elapsed time and CPU time

Same problem here, with another device.

Compared elapsed time with 1, 2 and 4 cores with CPU time and the elapsed time roughly doubles each time. As if there is only one processor crunching all the time, no matter the settings. Only if one processor alone is used is there no real difference between those times.

I tested with everything on 100 % (CPU, RAM, storage) and had the device plugged in all the time. As the processors is rather low-cost I assumed the cores had to share some resources which made them dependend from each other buth now I am not so sure anymore.

Device: Acer Iconia A1,
CPU: A7-quad (MT8125)
RAM: 1 GB
[Aug 4, 2013 8:49:52 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Huge difference between Elapsed time and CPU time

Looking at the system requirements for Vina for FightAids at https://secure.worldcommunitygrid.org/help/viewTopic.do?shortName=minimumreq I see that each thread requires 100 MB of RAM. A smartphone with less than 200 MB can only run one thread at a time with any other threads suspended.
[Aug 4, 2013 11:21:10 PM]   Link   Report threatening or abusive post: please login first  Go to top 
ChrisR1964
Cruncher
Usa
Joined: Dec 5, 2005
Post Count: 17
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Huge difference between Elapsed time and CPU time

Looking at the system requirements for Vina for FightAids at https://secure.worldcommunitygrid.org/help/viewTopic.do?shortName=minimumreq I see that each thread requires 100 MB of RAM. A smartphone with less than 200 MB can only run one thread at a time with any other threads suspended.


I have a Galaxy Note 2 and if I'm reading the specs correctly I have 2GB of RAM and with 4 cores that would only require 400 MB. But yet I experience the same slowdowns that get worse the more cores I use. At least in my case I don't think it is a RAM issue.
[Aug 5, 2013 2:24:19 AM]   Link   Report threatening or abusive post: please login first  Go to top 
dskagcommunity
Senior Cruncher
Austria
Joined: May 10, 2011
Post Count: 219
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Huge difference between Elapsed time and CPU time

Nobody talked about phones with less then 200mb ram ^^ or do you mean free ram?

But whatever, boinc is showing only 26mb use and still 116mb ram free. Cant see that vina using extra threads in androids taskmanager like on a PC. but ok im ratter new at android ^^
----------------------------------------
http://www.research.dskag.at
Crunching for my Dog who had "good" Braincancer.


----------------------------------------
[Edit 3 times, last edit by dskagcommunity at Aug 5, 2013 8:24:47 AM]
[Aug 5, 2013 8:12:10 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Sgt.Joe
Ace Cruncher
USA
Joined: Jul 4, 2006
Post Count: 7846
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Huge difference between Elapsed time and CPU time

Looking at the system requirements for Vina for FightAids at https://secure.worldcommunitygrid.org/help/viewTopic.do?shortName=minimumreq I see that each thread requires 100 MB of RAM. A smartphone with less than 200 MB can only run one thread at a time with any other threads suspended.


I have a Galaxy Note 2 and if I'm reading the specs correctly I have 2GB of RAM and with 4 cores that would only require 400 MB. But yet I experience the same slowdowns that get worse the more cores I use. At least in my case I don't think it is a RAM issue.

That implies there is a some different bottleneck in how these devices move data internally. Someone who is familiar with the architecture of how data flows in these devices should be able to shed some light on this. I am speculating maybe a single buss for all the cores or all the cores are fighting to use the same limited number of registers.
Cheers
----------------------------------------
Sgt. Joe
*Minnesota Crunchers*
[Aug 5, 2013 11:27:31 AM]   Link   Report threatening or abusive post: please login first  Go to top 
dskagcommunity
Senior Cruncher
Austria
Joined: May 10, 2011
Post Count: 219
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Huge difference between Elapsed time and CPU time

Then the advertisment of 4 cores in a device is getting more and more a marketingthing only it seems ^^
----------------------------------------
http://www.research.dskag.at
Crunching for my Dog who had "good" Braincancer.


[Aug 5, 2013 1:55:28 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Randzo
Senior Cruncher
Slovakia
Joined: Jan 10, 2008
Post Count: 339
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Huge difference between Elapsed time and CPU time

Hello,

thank you for your opinions and ideas where the issue could be.

@lawrence I have 832 MB RAM and I always have over 100 MB free when running the science apps.

For sure there is some bottleneck but I have no idea where it can be I am not very skilled with Android and mobile devices.

I will try to disable Power saving and will share the results after some tests.

Could please somebody try it with me so the results won't be so biased?

Btw. I'm not complaining I just would like to know and maybe improve performance of my Android device for the research.
[Aug 5, 2013 6:55:22 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Posts: 19   Pages: 2   [ 1 2 | Next Page ]
[ Jump to Last Post ]
Post new Thread