Welcome to MilkyWay@home

Micromanaging CPU vs GPU Workunit Limits

Message boards : News : Micromanaging CPU vs GPU Workunit Limits
Message board moderation

To post messages, you must log in.

Previous · 1 · 2

AuthorMessage
Profile mikey
Avatar

Send message
Joined: 8 May 09
Posts: 3339
Credit: 524,010,781
RAC: 0
Message 68061 - Posted: 24 Jan 2019, 21:58:04 UTC - in response to Message 68058.  

You do know you can turn each kind off individually in the settings...right?

Run only the selected applications MilkyWay@Home: yes
MilkyWay@Home N-Body Simulation: no


But that doesn't select between single-core and multi-core N-Body, does it?


Yes it does, if you unselect the "N-Body Simulation" units you won't get any more multi-core wu's, on the other hand if you unselect the "MilkyWay@Home" units then you won't get any more single core wu's.
ID: 68061 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Jim1348

Send message
Joined: 9 Jul 17
Posts: 100
Credit: 16,967,906
RAC: 0
Message 68062 - Posted: 25 Jan 2019, 2:29:15 UTC - in response to Message 68061.  

Yes it does, if you unselect the "N-Body Simulation" units you won't get any more multi-core wu's, on the other hand if you unselect the "MilkyWay@Home" units then you won't get any more single core wu's.

OK, it looks like I had both selected. I will try just the N-Body. Thanks.
ID: 68062 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
DAF

Send message
Joined: 7 Jan 16
Posts: 4
Credit: 16,016,591
RAC: 0
Message 68063 - Posted: 25 Jan 2019, 5:25:28 UTC

cpu and gpu is too much wats...need to choose something one.
ID: 68063 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Link
Avatar

Send message
Joined: 19 Jul 10
Posts: 623
Credit: 19,255,064
RAC: 6
Message 68064 - Posted: 25 Jan 2019, 8:26:21 UTC - in response to Message 68061.  
Last modified: 25 Jan 2019, 8:27:24 UTC

You do know you can turn each kind off individually in the settings...right?

Run only the selected applications MilkyWay@Home: yes
MilkyWay@Home N-Body Simulation: no


But that doesn't select between single-core and multi-core N-Body, does it?


Yes it does, if you unselect the "N-Body Simulation" units you won't get any more multi-core wu's, on the other hand if you unselect the "MilkyWay@Home" units then you won't get any more single core wu's.

No, it does not. With "MilkyWay@Home: yes", you will get single core MilkyWay@Home WUs (they are all single core), but not single core MilkyWay@Home N-Body Simulation WUs.
ID: 68064 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Link
Avatar

Send message
Joined: 19 Jul 10
Posts: 623
Credit: 19,255,064
RAC: 6
Message 68065 - Posted: 25 Jan 2019, 8:30:01 UTC - in response to Message 68063.  

cpu and gpu is too much wats...need to choose something one.

You can choose that already in your MilkyWay@Home preferences.
ID: 68065 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 8 May 09
Posts: 3339
Credit: 524,010,781
RAC: 0
Message 68066 - Posted: 25 Jan 2019, 11:34:15 UTC - in response to Message 68064.  

You do know you can turn each kind off individually in the settings...right?

Run only the selected applications MilkyWay@Home: yes
MilkyWay@Home N-Body Simulation: no


But that doesn't select between single-core and multi-core N-Body, does it?


Yes it does, if you unselect the "N-Body Simulation" units you won't get any more multi-core wu's, on the other hand if you unselect the "MilkyWay@Home" units then you won't get any more single core wu's.


No, it does not. With "MilkyWay@Home: yes", you will get single core MilkyWay@Home WUs (they are all single core), but not single core MilkyWay@Home N-Body Simulation WUs.


By definition here at MW N-Body wu's are multi-core wu's.
ID: 68066 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Link
Avatar

Send message
Joined: 19 Jul 10
Posts: 623
Credit: 19,255,064
RAC: 6
Message 68067 - Posted: 26 Jan 2019, 11:40:12 UTC - in response to Message 68066.  

By definition here at MW N-Body wu's are multi-core wu's.

No, there's also a single core N-Body application, see applications page.
ID: 68067 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 8 May 09
Posts: 3339
Credit: 524,010,781
RAC: 0
Message 68068 - Posted: 26 Jan 2019, 12:30:59 UTC - in response to Message 68067.  

By definition here at MW N-Body wu's are multi-core wu's.


No, there's also a single core N-Body application, see applications page.


Hmm I did not know that.... Thanks!!!
ID: 68068 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Dylan

Send message
Joined: 27 Dec 12
Posts: 2
Credit: 3,150,504
RAC: 0
Message 68070 - Posted: 26 Jan 2019, 21:30:46 UTC - in response to Message 68063.  

cpu and gpu is too much wats...need to choose something one.


Always GPU. Way more processing power for the inputted energy. Downside is it is harder to code applications for it.
ID: 68070 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
bluestang

Send message
Joined: 13 Oct 16
Posts: 112
Credit: 1,174,293,644
RAC: 0
Message 68075 - Posted: 28 Jan 2019, 14:36:22 UTC

Another 4+ hour "Server can't open database" issue again. Please increase our WU limits.

Also, how about changing the scheduled communication time between request from 1m30s to something higher? Won't that help with the amount of hits the server is also taking on? If we can have more WU downloaded, it doesn't need to try and connect every 1m30s.
ID: 68075 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
bluestang

Send message
Joined: 13 Oct 16
Posts: 112
Credit: 1,174,293,644
RAC: 0
Message 68109 - Posted: 7 Feb 2019, 20:19:58 UTC

So how about doubling the 600 limit for GPUs to 1200?
ID: 68109 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
jo1252

Send message
Joined: 4 Aug 09
Posts: 1
Credit: 6,160,894
RAC: 0
Message 68183 - Posted: 23 Feb 2019, 17:22:46 UTC

Here is the thing happening with the database flickering between on and off everyday for many hours at a time.

With 200 WUs per GPU, I run out of work in <2 hours.
The database usually goes offline for longer than this amount of time, putting effectively my computer on a crunching break multiple times per day.

I would strongly encourage you to either improve the stability of your servers, increase the limit of GPU WUs we can have stored to a much higher amount (600WUs per GPU, for example). Another suggestion would be to increase the work you put in each WU. That way they would run for longer on GPU and the lower limit you now have in place would be adequate. If your database gets crunched by too many requests, maybe you could implement a higher time between requests, now it is every 90 seconds, what about every 10 minutes or something higher to reduce the load on your end.

I really hope you will improve this situation.
ID: 68183 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Jim1348

Send message
Joined: 9 Jul 17
Posts: 100
Credit: 16,967,906
RAC: 0
Message 68184 - Posted: 24 Feb 2019, 2:50:17 UTC - in response to Message 68183.  

I really hope you will improve this situation.

I have concluded that if they are smart enough to map the galaxy, then they must be smart enough to know that by limiting the work units, they will lose the fastest machines to other projects (especially in view of all the server outages).

My guess is that they are getting enough work done with the slower machines to satisfy their needs. It may take time to analyze the results, for example, which can't be sped up further. So if that does not work for you, probably finding another project is the best solution. They all have their limits somewhere; there is not an infinite supply of work. We just need to move around to accommodate what they can supply at any given time.
ID: 68184 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 8 May 09
Posts: 3339
Credit: 524,010,781
RAC: 0
Message 68185 - Posted: 24 Feb 2019, 12:04:04 UTC - in response to Message 68184.  

I really hope you will improve this situation.

I have concluded that if they are smart enough to map the galaxy, then they must be smart enough to know that by limiting the work units, they will lose the fastest machines to other projects (especially in view of all the server outages).

My guess is that they are getting enough work done with the slower machines to satisfy their needs. It may take time to analyze the results, for example, which can't be sped up further. So if that does not work for you, probably finding another project is the best solution. They all have their limits somewhere; there is not an infinite supply of work. We just need to move around to accommodate what they can supply at any given time.


I agree the new "Do You Feel Lucky? (GFN World Record))" workunits at PrimeGrid, suggested for high end gpu's only, takes about 2 days on a 1080Ti and you get almost 1 million credits for each one. And they will make as many as they have people willing to crunch them. For cpu workunits they have some that take over 100 days!!
ID: 68185 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Jake Weiss
Volunteer moderator
Project developer
Project tester
Project scientist

Send message
Joined: 25 Feb 13
Posts: 580
Credit: 94,200,158
RAC: 0
Message 68196 - Posted: 1 Mar 2019, 19:48:41 UTC

Hey Everyone,

Sorry that things have been bad for so long. I just want to let you know that we have ordered a new server and it is on the way. Hopefully it will be here in a couple weeks and it should further relieve some of the database issues we have been experiencing.

Best,

Jake
ID: 68196 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
bluestang

Send message
Joined: 13 Oct 16
Posts: 112
Credit: 1,174,293,644
RAC: 0
Message 68197 - Posted: 4 Mar 2019, 16:41:41 UTC

Good news indeed Jake. Congrats!
ID: 68197 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
neofob

Send message
Joined: 4 Mar 18
Posts: 23
Credit: 265,141,401
RAC: 12,761
Message 68231 - Posted: 8 Mar 2019, 3:15:01 UTC

I wonder if there is a way to set the number of cached WU based on the rate of processed WU in the past 1-3 days? This might be better than basing on the number of GPU.
ID: 68231 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Hurr1cane78

Send message
Joined: 7 May 14
Posts: 57
Credit: 206,540,646
RAC: 34
Message 69783 - Posted: 10 May 2020, 8:37:56 UTC

hi all made vid on youtube for multiple instances instruction's and at full load on a Radeon VII
RADEON VII GIGABYTE// 3 Instances_ Milkyway@home WUs BOINC_ 3_instances
https://www.youtube.com/watch?v=4xKy9wGKmz4
all the best and welcome to earth
ID: 69783 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Previous · 1 · 2

Message boards : News : Micromanaging CPU vs GPU Workunit Limits

©2024 Astroinformatics Group