PrimeGrid
Please visit donation page to help the project cover running costs for this month

Toggle Menu

Join PrimeGrid

Returning Participants

Community

Leader Boards

Results

Other

drummers-lowrise
11) Message boards : Generalized Fermat Prime Search : Multiple Genefer GPU Apps on a single card. (Message 119593)
Posted 854 days ago by AeneasProject donor
Thanks Michael.
12) Message boards : Generalized Fermat Prime Search : Multiple Genefer GPU Apps on a single card. (Message 119589)
Posted 854 days ago by AeneasProject donor
Im having issues with this. Could someone point out what im doing wrong.

I want to crunch on gfn16 on my cpu

<app_config>

<app_version>
<app_name>genefer16</app_name>
<plan_class>cpuGFN16</plan_class>
<cmdline>-t 6</cmdline>
<avg_ncpus>1</avg_ncpus>
</app_version>

</app_config>

Should this attempt to use 6 processors?
13) Message boards : Problems and Help : No usable GPUs found (Message 119560)
Posted 856 days ago by AeneasProject donor
I've got the same issue on sepdk. If I disconnect my rdp session will it start using the GPU's?

14) Message boards : Number crunching : Free-DC changes (Message 118542)
Posted 898 days ago by AeneasProject donor
GCW LLR link works now. Thank you very much.
15) Message boards : Number crunching : Free-DC changes (Message 118486)
Posted 900 days ago by AeneasProject donor
New look is awesome. I note that the GCW LLR link is broken though.
16) Message boards : Project Staging Area : Where would the PRPNET projects rank? (Message 118427)
Posted 901 days ago by AeneasProject donor
Hi,

Where would the PRPNET projects rank if a prime was found in one of them?
17) Message boards : General discussion : Favourite Prime Number. (Message 118423)
Posted 902 days ago by AeneasProject donor
Well aside from the FPS i've found I like 2.
18) Message boards : Problems and Help : Another Question about multi-threading (Message 116837)
Posted 964 days ago by AeneasProject donor
By moving to 2 * 4 it looks like I'll be able to do 50% more if the run times continue to hold up.

eg
1 * 8 9357 secs on average per WU ~ 64 per week
2 * 4 140633 secs on average per WU ~ 86 per week

If my admittedly hazy math is correct.
19) Message boards : Problems and Help : Another Question about multi-threading (Message 116783)
Posted 965 days ago by AeneasProject donor
Thanks Michael. I'll have a play around and see.

i assume to get two going



<cmdline>-t 8</cmdline> I change this to 4
<avg_ncpus>1</avg_ncpus> and this to 2


As well as upping the % of processors from 12.5% to 25%
20) Message boards : Problems and Help : Another Question about multi-threading (Message 116732)
Posted 967 days ago by AeneasProject donor
Hi,

I have a host 909979 which has 2 sockets with four cores. Currently im running it with hyper threading off. ie im getting 8 logical processors crunching rather than 16.

I have an app_config.xml that looks like:

<app_config>

<app_version>
<app_name>llr321</app_name>
<cmdline>-t 8</cmdline>
<avg_ncpus>1</avg_ncpus>
</app_version>

</app_config>

The machine has 64 gb of ram of 2133mhz RAM arranged in two banks of 4 * 8gb.

1) Should I be turning hyper threading on?

2) Is there a point where multi-threading becomes less efficient and for example maybe I should be doing two tasks at once with 4 processors?

Any help would be gratefully received.





Next 10 posts
[Return to PrimeGrid main page]
DNS Powered by DNSEXIT.COM
Copyright © 2005 - 2020 Rytis Slatkevičius (contact) and PrimeGrid community. Server load 0.00, 0.01, 0.00
Generated 25 Nov 2020 | 0:44:14 UTC