20220509, 18:20  #34  
Sep 2002
Database er0rr
1000010111011_{2} Posts 
Quote:
Perhaps EdH can give us a synopsis on how to run client boxes. Last fiddled with by paulunderwood on 20220509 at 18:23 

20220509, 20:04  #35  
"Ed Hall"
Dec 2009
Adirondack Mtns
3^{3}×181 Posts 
Quote:
Basically, I have an exact same user account on all the machines, with an exact same directory tree, and all machines are key linked via ssh with every other machine. This linking of all machines to all others is important. ecmpi is compiled on each machine, and the hostfile lists the localhost slots= and the user@IP slots= for each client. I can get more detailed later. Let me look over the rest of the thread. Or, I'll try to answer more questions. 

20220509, 20:35  #36 
"Ed Hall"
Dec 2009
Adirondack Mtns
3^{3}·181 Posts 
I'm not familiar enough with all the required libraries to implement this, yet. But, it appears that others are running openmpi and xilman knows how to incorporate LAN connected clients (which I already suspected). I'll address further questions, if any arise.

20220509, 20:41  #37 
"Serge"
Mar 2008
Phi(4,2^7658614+1)/2
3×3,313 Posts 
I found an interesting feature in this implementation  it doesn't backtrack. Could be something that is left for later work. This has a downside of sometimes taking more time in one single step than all the others combined.
In proving a 12k digit number, one step took 10 hrs (already near 10k digit size, i.e. similar to Greg's number above which took 20hrs in all, and on similar node) and then proceeded fast. This is deterministic, too  when repeated again from that step, it took ~10 hours again. I sent the debug case input to Andreas. 
20220510, 14:10  #38 
Jul 2003
So Cal
2,473 Posts 
Certification of a 15k digit prime, 2^498313637, took 29.7 hours using 12x Xeon E52670 v2 cpus with a total of 120 cores and 240 hyperthreads. The longest step of the second phase took 6.3 hours to complete.
It appears that idle processes near the end of the second phase continue to use processor time. I will next try a run with one process per core rather than two. 
20220510, 15:24  #39 
Jan 2007
Germany
2^{2}×3×43 Posts 
Wow , 15k in 30h ..that is fast !
So I have 3 milestone 20k numbers for testing ! Smallest 20000 digit prime and prime twins. 10^19999+110949 10^19999+1514722609+d, d=0,2 Should be done, the prover extended this page... https://pzktupel.de/smx000.html regards 
20220510, 15:57  #40  
Sep 2002
Database er0rr
4,283 Posts 
questions about mpirun ecppmpi
Quote:
As far as I can tell, the clients require passwordless ssh keys along with identical project directories and an installation of mpi. Do the clients require ecppmpi too? Last fiddled with by paulunderwood on 20220510 at 16:04 

20220510, 16:52  #41  
Jul 2003
So Cal
2,473 Posts 
Quote:
No No Quote:
Yes, each compiled for the processor on which it will be run. 

20220510, 18:16  #42  
Jul 2003
So Cal
2,473 Posts 
Quote:


20220510, 18:36  #43  
Bamboozled!
"๐บ๐๐ท๐ท๐ญ"
May 2003
Down not across
11500_{10} Posts 
Quote:
Looks like my choice of 1 process per core was (accidentally) correct. Still wondering whether n+1 on a ncore machine might be a better choice to soak up some of the hyperthreading and have n workers alongside the single master. Next time, perhaps. 

20220511, 08:07  #44 
"Oliver"
Sep 2017
Porta Westfalica, DE
2177_{8} Posts 
On a fixed system, fastECPP was around double the speed of Primo for a ~10k digit prime.

Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
For which types of primes is GPU primality test software available?  bur  GPU Computing  6  20200828 06:20 
Fastest software for Mersenne primality test?  JonathanM  Information & Answers  25  20200616 02:47 
APRCL as primality proof  f1pokerspeed  FactorDB  14  20140109 21:06 
Proof of Primality Test for Fermat Numbers  princeps  Math  15  20120402 21:49 
PRIMALITY PROOF for Wagstaff numbers!  AntonVrba  Math  96  20090225 10:37 