20200310, 18:17  #12 
"Carlos Pinho"
Oct 2011
Milton Keynes, UK
3×1,663 Posts 
HT helps a lot on LA, at least for me.

20200312, 23:36  #13  
"Curtis"
Feb 2005
Riverside, CA
11712_{8} Posts 
Quote:
VBITS=128 on otherwise idle machine. ETA after 1% of job: 6threaded 14hr 34 min 12threads 8 hr 26 min 18threads 9 hr 15 min 24threads 8 hr 27 min These times look rather slow; I just installed the extra 32GB memory today, so perhaps filling all 8 slots slows memory access a bunch. Some time I'll remove the original 16GB and see if 4 sticks is faster than 8. 

20200427, 12:16  #14  
Aug 2002
2^{2}×2,087 Posts 
Quote:
Note that we only count the LA phase in our calculations. 

20200427, 22:12  #15  
Aug 2002
2^{2}·2,087 Posts 
Quote:
12 = 8h04m50s 20 = 8h32m31s 24 = 7h59m33s 

20200807, 16:31  #16 
Aug 2002
209C_{16} Posts 
CPU = i78565U
RAM = 2×16GB DDR42400 CMD = ./msieve v nc t 8 LA = 47884s 
20200808, 17:00  #17 
Aug 2002
2^{2}×2,087 Posts 
CPU = i78565U
RAM = 2×16GB DDR42400 CMD = ./msieve v nc t 4 LA = 51662s 
20200812, 17:59  #18 
Aug 2002
2^{2}×2,087 Posts 
CPU = 3950X
RAM = 2×8GB DDR43666 CMD = ./msieve v nc t 16 LA = 27180s 
20200814, 07:53  #19 
"Beschorner Kurt"
Jul 2016
Germany
14_{16} Posts 
In my experience, the throughput depends on an additionally running program (e.g. gmpecm)
machine: i77820X  8 cores + HT matrix: 49M * 49M memory: 64 GB msieve .... t16 solo ~ 55% (power according task manager) msieve .... t16 and gmpecm (prior: low) ~ 78% " With msieve + mprime/Prime95 the effectiveness is a litle lower Kurt 
20200814, 18:06  #20  
Jul 2003
So Cal
3^{2}×13×19 Posts 
Quote:
mpirun np 2 msieve nc2 1,2 v t 20 

20200815, 07:23  #21 
Jul 2003
So Cal
3^{2}×13×19 Posts 
Here's a bench using compute nodes with one Xeon E52650 v4 Broadwell cpu with 12cores, 24 threads.
1 node 7h 40m 2 nodes 2h 45m 4 nodes 1h 35m 8 nodes 1h 10m Not sure why the time for one node is so high compared to the others? Perhaps something fitting into the cache with the smaller matrices on each node? 
20200815, 15:58  #22  
"Curtis"
Feb 2005
Riverside, CA
13CA_{16} Posts 
Quote:


Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
PFGW benchmarking  carpetpool  Hardware  4  20190930 20:06 
Looking for benchmarking help with a Phenom or PhenomII X6  mrolle  Software  25  20120314 14:15 
GMP 5.0.1 vs GMP 4.1.4 benchmarking  unconnected  GMPECM  5  20110403 16:16 
Benchmarking dualCPU machines  garo  Software  2  20100927 20:33 
Benchmarking challenge!  Xyzzy  Software  17  20030826 15:43 