value for wetstone benchmark absurdly high, Ubuntu 20.04, Boinc 7.16.6. Will this effect credit/work received?

Message boards : Number crunching : value for wetstone benchmark absurdly high, Ubuntu 20.04, Boinc 7.16.6. Will this effect credit/work received?

To post messages, you must log in.

AuthorMessage
wolfman1360

Send message
Joined: 18 Feb 17
Posts: 72
Credit: 18,450,036
RAC: 0
Message 102442 - Posted: 24 Aug 2021, 6:07:42 UTC

Hello,
I just noticed this, and not sure if this will make a difference in credit received by my machines on this project. The wetstone value seems incredibly high.
I just ran a benchmark via Boinc. Intel xeon e5-2680v1.
8/24/2021 1:01:14 AM | | Running CPU benchmarks
8/24/2021 1:01:45 AM | | Benchmark results:
8/24/2021 1:01:45 AM | | Number of CPUs: 32
8/24/2021 1:01:45 AM | | 3817 floating point MIPS (Whetstone) per CPU
8/24/2021 1:01:45 AM | | 56045 integer MIPS (Dhrystone) per CPU
Anyone else experiencing this? Should I just ignore it?
thanks!
ID: 102442 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Grant (SSSF)

Send message
Joined: 28 Mar 20
Posts: 1491
Credit: 14,702,529
RAC: 15,519
Message 102443 - Posted: 24 Aug 2021, 6:31:40 UTC
Last modified: 24 Aug 2021, 6:32:30 UTC

Rosetta uses the benchmarks to determine Credit. However under certain conditions it also uses BOINC's Credit New Mechanism.
The end result- initial Credit granted may be high/low, but it will eventually settle down to a more normal level if the benchmark values are too high/low (when work is available again).

To me (not that i've paid much attention too them, but having a look at a few other systems values) it doesn't appear to be an unreasonable value for that CPU.
Grant
Darwin NT
ID: 102443 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile lazyacevw

Send message
Joined: 18 Mar 20
Posts: 12
Credit: 93,576,463
RAC: 0
Message 102447 - Posted: 24 Aug 2021, 10:17:35 UTC - in response to Message 102442.  
Last modified: 24 Aug 2021, 10:18:03 UTC

Looks about right to me but I haven't paid too much attention to the benchmarks boinc uses. Here is what I see on a few of my systems (whetstone):

E5-1620 - 4470 FP MIPS
3970X - 7221 FP MIPS
3950X - 6155 FP MIPS
3900X - 7262 FP MIPS

The important thing is to run the benchmark once and let boinc do all of the rest.
ID: 102447 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
wolfman1360

Send message
Joined: 18 Feb 17
Posts: 72
Credit: 18,450,036
RAC: 0
Message 102450 - Posted: 24 Aug 2021, 21:32:47 UTC

Thank you for that info.
I'll just let Boinc and the project worry about it. I just wanted to make sure nothing was amiss.
Here's hoping more work appears.
ID: 102450 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Sid Celery

Send message
Joined: 11 Feb 08
Posts: 2000
Credit: 38,573,281
RAC: 17,120
Message 102453 - Posted: 25 Aug 2021, 5:37:07 UTC - in response to Message 102442.  

I just noticed this, and not sure if this will make a difference in credit received by my machines on this project. The whetstone value seems incredibly high....

...Should I just ignore it?

Yes, you should do your best to have them removed... lol

Bank it and hope no-one tries to fix it. Blimey...
ID: 102453 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote

Message boards : Number crunching : value for wetstone benchmark absurdly high, Ubuntu 20.04, Boinc 7.16.6. Will this effect credit/work received?



©2024 University of Washington
https://www.bakerlab.org