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: 16
Posts: 16   Pages: 2   [ Previous Page | 1 2 ]
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 4004 times and has 15 replies Next Thread
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Is MIP more disruptive than other WCG projects?


I have a Ryzen 5 1600, with 16GB of RAM (two 8GB DDR4 2667 sticks). I have a mild overclock with all 6 cores running at 3.65GHz. I run 6 cores/12 threads at 100% 24/7. Typically 75 - 80% of my 12 threads are running MIP1 jobs. Based on recent results returned, it takes on average 2.18 hrs for this PC to complete a MIP task. What is the average time to complete a MIP task on your R7 1700? (I'm guessing Linux is much faster than Win10, but you have only 4GB of Ram, so curious). FWIW, I haven't noticed any slowdown in web browsing or in MS Office applications while all of these BOINC related jobs are running on this PC.

I also run an i5 4670K at 4.0GHz with 16GB of DDR3 RAM. I run 4 cores/4 threads at 100% 24/7. Typically 75% of my 4 threads are running MIP1 jobs. Based on recent results returned, it takes on average 1.17 hrs for this PC to complete a MIP task. I do notice a slow down in MS Office/browsing when running BOINC at 100% on this PC.


My Ryzen 1700 @3.2 GHz averaged 1.55 hrs for the last few days, with normally 3 of the 16 threads running MIP. The only 4 GB RAM of my machine should not be an issue here, Lubuntu itself uses only ~300 MB and except from WCG nothing is running on the machine, typically only 2 GB RAM are used overall. Windows vs Linux should also be no large difference in MIP, only at the VINA projects (OET, Zika, SCC, FAH1) Linux has a large speed advantage.
I would guess your MIP completion times will improve noticably, if you run fewer of them together with other projects. My Ryzen 1700 runs 3 MIP units without any measurable performance impact, 4 with a slight impact, with 5 and more units the slowdown becomes significant. As your Ryzen 1600 has 16 MB L3 cache, the same as my 1700, I assume this should be the same on your machine.
[May 3, 2018 3:00:42 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Macromancer
Veteran Cruncher
United States
Joined: Sep 6, 2016
Post Count: 994
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Is MIP more disruptive than other WCG projects?


My Ryzen 1700 @3.2 GHz averaged 1.55 hrs for the last few days, with normally 3 of the 16 threads running MIP.


Super. I added MCM to my project list last night, so I should have fewer MIP tasks on average going forward. I'll calculate an average run time for MIP tasks in a few days when I have enough data for a meaningful comparison. Based on a very small sample, I'm now getting an average of 1.82 hrs to complete a MIP task, so moving in the right direction.

I'd prefer to limit the MIP tasks via the config/xml file, but I couldn't find any of the directories mentioned in the BOINC documenation, i.e. my Win 10 BOINC directory structure looks quite a bit different from what I see in the documentation.
[May 3, 2018 4:46:12 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: Is MIP more disruptive than other WCG projects?

At my Windows 10 PC the relevant directory can be found at

C:\ProgramData\BOINC\projects\www.worldcommunitygrid.org

Remember the dir could be hidden, so either show hidden folder in the settings or enter this path with copy/paste.

In case there is no app_config.xml you must create one...

In case you still can't find the directory, maybe searching for "www.worldcommunitygrid.org" on your PC could help, to locate it.
[May 3, 2018 6:17:20 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Macromancer
Veteran Cruncher
United States
Joined: Sep 6, 2016
Post Count: 994
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Is MIP more disruptive than other WCG projects?

After about 36hrs of running while limiting the number of MIP tasks, I definitely observe that it takes less time on average to complete a MIP task:

i5 4670K @4.0 GHz: an 11% improvement (now 1.05 hrs versus 1.17 hrs previously)

R5 1600 @3.4 GHz: a 23% improvement (now 1.67 hrs versus 2.18 hrs previously). I had to lower the over clock from 3.6 to 3.4GHz on all cores after limiting the number of MIP tasks in order to keep CPU temps below 60C while running at 100%; 60C is just the arbitrary max CPU package temp I prefer.
[May 4, 2018 2:19:58 PM]   Link   Report threatening or abusive post: please login first  Go to top 
l_mckeon
Senior Cruncher
Joined: Oct 20, 2007
Post Count: 439
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Is MIP more disruptive than other WCG projects?

Another thing is that MIP may not be suitable for those on expensive wireless data plans. I've done a rough calculation and for my old 3.3GHz I5-2500, MIP will have used around 21-22GB of data to get a 90 day badge.
[May 25, 2018 7:45:05 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Jim1348
Veteran Cruncher
USA
Joined: Jul 13, 2009
Post Count: 1066
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Is MIP more disruptive than other WCG projects?

I use an app_config.xml file to limit MIP to two work units at a time on my Ryzen 1700 (Lubuntu 17.10.1, not overclocked). I did some tests several months ago, and found that more than two might lengthen run times, though I did not try to nail it down precisely. But I allow all other work units too, so I haven't had the "bunching up" problem yet.

The run times are presently estimated at 1 hour 23 minutes by BOINCTasks, though that can change.
[May 28, 2018 2:12:27 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Posts: 16   Pages: 2   [ Previous Page | 1 2 ]
[ Jump to Last Post ]
Post new Thread