The Hunt Is ON!

Come join the... uh... er... fold.

Moderators: just brew it!, farmpuma

Re: The Hunt Is ON!

Postposted on Sat Nov 26, 2011 12:31 am

But I average a good 10000+ ppd less with the non-bigadv WU's.
Ragnar Dan
Gerbil Elder
Silver subscriber
 
 
Posts: 5354
Joined: Sun Jan 20, 2002 7:00 pm

Re: The Hunt Is ON!

Postposted on Sat Nov 26, 2011 3:13 am

Ragnar Dan wrote:But I average a good 10000+ ppd less with the non-bigadv WU's.

I average less with bigadv WUs, though I have the i7-875K running at stock may be it is not fast enough?
Image
The Model M is not for the faint of heart. You either like them or hate them.

Gerbils unite! Fold for UnitedGerbilNation, team 2630.
Flying Fox
Gerbil God
 
Posts: 24289
Joined: Mon May 24, 2004 2:19 am

Re: The Hunt Is ON!

Postposted on Sat Nov 26, 2011 8:20 am

Ragnar Dan wrote:But I average a good 10000+ ppd less with the non-bigadv WU's.

Well, I did say "reasonably well" not "as well", since I wasn't sure how the performance compared. When was the last time you tried running SMP WUs with a passkey? I'm seeing increases of up to a factor of 8 over the "raw" point score (depending on the CPU), due to the early return bonus. My Phenom II 1090T (running stock) is capable of consistently pumping out 15-18K PPD.

Flying Fox wrote:I average less with bigadv WUs, though I have the i7-875K running at stock may be it is not fast enough?

So bigadv is not necessarily better in all cases...
(this space intentionally left blank)
just brew it!
Administrator
Gold subscriber
 
 
Posts: 37505
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: The Hunt Is ON!

Postposted on Sat Nov 26, 2011 4:37 pm

just brew it! wrote:
Ragnar Dan wrote:But I average a good 10000+ ppd less with the non-bigadv WU's.

Well, I did say "reasonably well" not "as well", since I wasn't sure how the performance compared. When was the last time you tried running SMP WUs with a passkey? I'm seeing increases of up to a factor of 8 over the "raw" point score (depending on the CPU), due to the early return bonus. My Phenom II 1090T (running stock) is capable of consistently pumping out 15-18K PPD.

I got them for quite a few days in a row when the bigadv WU's began running out. I got just one bigadv this week, all the rest have been "regular". With respect to your 1090T, it seems to have the single clock SSE instructions Intel intro'd in the C2D that previous AMD CPU's lacked, which helps a lot. Or so I guess, considering its speed.

just brew it! wrote:
Flying Fox wrote:I average less with bigadv WUs, though I have the i7-875K running at stock may be it is not fast enough?

So bigadv is not necessarily better in all cases...

He doesn't OC his enough. :)
Ragnar Dan
Gerbil Elder
Silver subscriber
 
 
Posts: 5354
Joined: Sun Jan 20, 2002 7:00 pm

Re: The Hunt Is ON!

Postposted on Sat Jan 21, 2012 12:54 pm

Gentlemen, I need some advice. My FAH farm has taken a hit last year with multiple machine failures.

The remaining machines have been getting low PPD for months now. I know BIGADV is not available for < 8 cores on Windows, but still the PPD seems very low compared to others with 2600K's and 970's.

3x Windows XP-32 bit PC's:

1) 2600K @ 4.2 Ghz - 8 Threads
2) i7-970 @ 3.5Ghz - 10 Threads
3) i7-920 @ 2.6Ghz - 8 Threads

All 3 PC's produce approximately the same PPD (depending on WU): 13-18k PPD. The PC's run a GPU or 2.

The OC could go a bit higher on them, but regardless, should I be looking at a move to Linux or am I getting what I should be out of this hardware?

- JP
JPinTO
Gerbil Team Leader
 
Posts: 240
Joined: Sat Jun 30, 2007 7:02 am
Location: Toronto, Ontario

Re: The Hunt Is ON!

Postposted on Sat Jan 21, 2012 2:53 pm

JPinTO wrote:Gentlemen, I need some advice. My FAH farm has taken a hit last year with multiple machine failures.

The remaining machines have been getting low PPD for months now. I know BIGADV is not available for < 8 cores on Windows, but still the PPD seems very low compared to others with 2600K's and 970's.

3x Windows XP-32 bit PC's:

1) 2600K @ 4.2 Ghz - 8 Threads
2) i7-970 @ 3.5Ghz - 10 Threads
3) i7-920 @ 2.6Ghz - 8 Threads

All 3 PC's produce approximately the same PPD (depending on WU): 13-18k PPD. The PC's run a GPU or 2.

The OC could go a bit higher on them, but regardless, should I be looking at a move to Linux or am I getting what I should be out of this hardware?

- JP

I have not been running my Linux folding VMs for a year or so since WinSMP has become somewhat hassle free, even under service mode. I suppose if LinuxSMP has the advantage again I can switch my own node to a VM. Anyone has recent hands-on knowledge about the ppd difference?
Image
The Model M is not for the faint of heart. You either like them or hate them.

Gerbils unite! Fold for UnitedGerbilNation, team 2630.
Flying Fox
Gerbil God
 
Posts: 24289
Joined: Mon May 24, 2004 2:19 am

Re: The Hunt Is ON!

Postposted on Sat Jan 21, 2012 3:27 pm

Unfortunately I've got zero recent experience with any of the folding clients on Intel CPUs or Windows. All of my folding for the past few years has been on AMD CPUs running Linux, so I really can't say whether your PPD is good, bad, or indifferent for your hardware.

My Phenom II X6 1090T does around 18K PPD running the Linux SMP client, if that helps (it probably doesn't...) :-?
(this space intentionally left blank)
just brew it!
Administrator
Gold subscriber
 
 
Posts: 37505
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: The Hunt Is ON!

Postposted on Sat Jan 21, 2012 5:48 pm

When I ran F@H in Linux under VMWare, I received a noticeable increase in output compared to Windows, though since they quit allowing users with fewer than 16 hardware threads to fold any kind of -bigadv WU's, I've quit most of my folding.

Here's an announcement indicating the fact that 16 threads is the minimum for -bigadv WU's, somewhat vaguely as per usual with the ingenius types at Stanford: http://foldingforum.org/viewtopic.php?f=24&t=20349

Also, Vijay Pande's blog mentioned this on November 14, 2011.
Ragnar Dan
Gerbil Elder
Silver subscriber
 
 
Posts: 5354
Joined: Sun Jan 20, 2002 7:00 pm

Re: The Hunt Is ON!

Postposted on Sat Jan 21, 2012 10:36 pm

Ragnar Dan wrote:When I ran F@H in Linux under VMWare, I received a noticeable increase in output compared to Windows, though since they quit allowing users with fewer than 16 hardware threads to fold any kind of -bigadv WU's, I've quit most of my folding.

So in your experience, current LinuxSMP, even without -bigadv, is now getting the upperhand (first it was LinuxSMP > WinSMP, then WinSMP got the upperhand even w/o bigadv, and now the tide has turned again?) over Windows in terms of points, enough to offset any VM overheads compared to running WinSMP as a service on a Windows host?
Image
The Model M is not for the faint of heart. You either like them or hate them.

Gerbils unite! Fold for UnitedGerbilNation, team 2630.
Flying Fox
Gerbil God
 
Posts: 24289
Joined: Mon May 24, 2004 2:19 am

Re: The Hunt Is ON!

Postposted on Mon Jan 23, 2012 7:48 pm

I wonder if the core count spoofing trick is worth doing, even though it's not "authorized"? If your machine can process the WU in the allotted time, then thats all that should matter. 6HTx4.0Ghz = 36Ghz. 2CPUx 8x 2.0Ghz is 16 cores, but still only 32Ghz.

- JP
JPinTO
Gerbil Team Leader
 
Posts: 240
Joined: Sat Jun 30, 2007 7:02 am
Location: Toronto, Ontario

Re: The Hunt Is ON!

Postposted on Mon Jan 23, 2012 11:24 pm

Flying Fox wrote:
Ragnar Dan wrote:When I ran F@H in Linux under VMWare, I received a noticeable increase in output compared to Windows[...]
So in your experience, current LinuxSMP, even without -bigadv, is now getting the upperhand (first it was LinuxSMP > WinSMP, then WinSMP got the upperhand even w/o bigadv, and now the tide has turned again?) over Windows in terms of points, enough to offset any VM overheads compared to running WinSMP as a service on a Windows host?

I do not have memory of any time when WinSMP was faster than Linux. I began running LinuxSMP in early 2009, and when -bigadv started it was Linux only at first IIRC, so I never ran Windows until later. The only gain from Windows was the ease of adding running a service compared to Linux. Though with VMWare server it wasn't too bad after getting help from notfred and, for me, reading the Ubuntu forum.

I saw your post last night and decided to see how a comparison went, though the definitive way to do it would be for me to learn how to read a Windows network share from VMWare'd Linux so that I could run N frames in one system and then restart it in the other and compare results. But it appears things are more than 10 seconds per frame slower on similar WU's. Here are 2 P7504's I ran:

Code: Select all
[06:47:06] Project: 7504 (Run 6, Clone 223, Gen 54)
[06:47:06]
[06:47:06] Assembly optimizations on if available.
[06:47:06] Entering M.D.
[06:47:12] Mapping NT from 8 to 8
[06:47:13] Completed 0 out of 500000 steps  (0%)
[06:51:28] Completed 5000 out of 500000 steps  (1%)
[06:55:39] Completed 10000 out of 500000 steps  (2%)
[06:59:53] Completed 15000 out of 500000 steps  (3%)
[07:04:05] Completed 20000 out of 500000 steps  (4%)
[07:08:16] Completed 25000 out of 500000 steps  (5%)
[07:12:29] Completed 30000 out of 500000 steps  (6%)
[07:16:41] Completed 35000 out of 500000 steps  (7%)
[07:20:54] Completed 40000 out of 500000 steps  (8%)
[07:25:05] Completed 45000 out of 500000 steps  (9%)
[07:29:20] Completed 50000 out of 500000 steps  (10%)
[07:33:35] Completed 55000 out of 500000 steps  (11%)
[07:37:44] Completed 60000 out of 500000 steps  (12%)
[07:41:57] Completed 65000 out of 500000 steps  (13%)
[07:46:10] Completed 70000 out of 500000 steps  (14%)
[07:50:25] Completed 75000 out of 500000 steps  (15%)
[07:54:35] Completed 80000 out of 500000 steps  (16%)
[07:58:53] Completed 85000 out of 500000 steps  (17%)
[08:03:04] Completed 90000 out of 500000 steps  (18%)
[08:07:15] Completed 95000 out of 500000 steps  (19%)
[08:11:26] Completed 100000 out of 500000 steps  (20%)
[08:15:39] Completed 105000 out of 500000 steps  (21%)
[08:19:49] Completed 110000 out of 500000 steps  (22%)
[08:24:01] Completed 115000 out of 500000 steps  (23%)
[08:28:11] Completed 120000 out of 500000 steps  (24%)
[08:32:23] Completed 125000 out of 500000 steps  (25%)
[08:36:34] Completed 130000 out of 500000 steps  (26%)
[08:40:45] Completed 135000 out of 500000 steps  (27%)
[08:44:58] Completed 140000 out of 500000 steps  (28%)
[08:49:09] Completed 145000 out of 500000 steps  (29%)
[08:53:22] Completed 150000 out of 500000 steps  (30%)
[08:57:32] Completed 155000 out of 500000 steps  (31%)
[09:01:33] Completed 160000 out of 500000 steps  (32%)
[09:05:46] Completed 165000 out of 500000 steps  (33%)
[09:09:57] Completed 170000 out of 500000 steps  (34%)
[09:14:10] Completed 175000 out of 500000 steps  (35%)
[09:18:20] Completed 180000 out of 500000 steps  (36%)
[09:22:31] Completed 185000 out of 500000 steps  (37%)
[09:26:45] Completed 190000 out of 500000 steps  (38%)
[09:30:58] Completed 195000 out of 500000 steps  (39%)
[09:35:06] Completed 200000 out of 500000 steps  (40%)
[09:39:17] Completed 205000 out of 500000 steps  (41%)
[09:43:28] Completed 210000 out of 500000 steps  (42%)
[09:47:38] Completed 215000 out of 500000 steps  (43%)
[09:51:51] Completed 220000 out of 500000 steps  (44%)
[09:56:01] Completed 225000 out of 500000 steps  (45%)
[10:00:12] Completed 230000 out of 500000 steps  (46%)
[10:04:26] Completed 235000 out of 500000 steps  (47%)
[10:08:39] Completed 240000 out of 500000 steps  (48%)
[10:12:49] Completed 245000 out of 500000 steps  (49%)
[10:17:03] Completed 250000 out of 500000 steps  (50%)
[10:21:19] Completed 255000 out of 500000 steps  (51%)
[10:25:32] Completed 260000 out of 500000 steps  (52%)
[10:29:45] Completed 265000 out of 500000 steps  (53%)
[10:33:59] Completed 270000 out of 500000 steps  (54%)
[10:38:13] Completed 275000 out of 500000 steps  (55%)
[10:42:26] Completed 280000 out of 500000 steps  (56%)
[10:46:39] Completed 285000 out of 500000 steps  (57%)
[10:50:53] Completed 290000 out of 500000 steps  (58%)
[10:55:05] Completed 295000 out of 500000 steps  (59%)
[10:59:17] Completed 300000 out of 500000 steps  (60%)
[11:03:30] Completed 305000 out of 500000 steps  (61%)
[11:07:42] Completed 310000 out of 500000 steps  (62%)
[11:11:57] Completed 315000 out of 500000 steps  (63%)
[11:16:08] Completed 320000 out of 500000 steps  (64%)
[11:20:21] Completed 325000 out of 500000 steps  (65%)
[11:24:32] Completed 330000 out of 500000 steps  (66%)
[11:28:46] Completed 335000 out of 500000 steps  (67%)
[11:33:01] Completed 340000 out of 500000 steps  (68%)
[11:37:11] Completed 345000 out of 500000 steps  (69%)
[11:41:27] Completed 350000 out of 500000 steps  (70%)
[11:45:39] Completed 355000 out of 500000 steps  (71%)
[11:49:54] Completed 360000 out of 500000 steps  (72%)
[11:54:05] Completed 365000 out of 500000 steps  (73%)
[11:58:17] Completed 370000 out of 500000 steps  (74%)
[12:02:30] Completed 375000 out of 500000 steps  (75%)
[12:06:43] Completed 380000 out of 500000 steps  (76%)
[12:10:56] Completed 385000 out of 500000 steps  (77%)
[12:15:08] Completed 390000 out of 500000 steps  (78%)
[12:19:19] Completed 395000 out of 500000 steps  (79%)
[12:23:31] Completed 400000 out of 500000 steps  (80%)
[12:27:42] Completed 405000 out of 500000 steps  (81%)
[12:31:52] Completed 410000 out of 500000 steps  (82%)
[12:36:00] Completed 415000 out of 500000 steps  (83%)
[12:40:17] Completed 420000 out of 500000 steps  (84%)
[12:44:32] Completed 425000 out of 500000 steps  (85%)
[12:46:58] - Autosending finished units... [January 23 12:46:58 UTC]
[12:46:58] Trying to send all finished work units
[12:46:58] + No unsent completed units remaining.
[12:46:58] - Autosend completed
[12:48:47] Completed 430000 out of 500000 steps  (86%)
[12:53:03] Completed 435000 out of 500000 steps  (87%)
[12:57:16] Completed 440000 out of 500000 steps  (88%)
[13:01:31] Completed 445000 out of 500000 steps  (89%)
[13:05:47] Completed 450000 out of 500000 steps  (90%)
[13:10:02] Completed 455000 out of 500000 steps  (91%)
[13:14:19] Completed 460000 out of 500000 steps  (92%)
[13:18:34] Completed 465000 out of 500000 steps  (93%)
[13:22:51] Completed 470000 out of 500000 steps  (94%)
[13:27:11] Completed 475000 out of 500000 steps  (95%)
[13:31:26] Completed 480000 out of 500000 steps  (96%)
[13:35:41] Completed 485000 out of 500000 steps  (97%)
[13:39:57] Completed 490000 out of 500000 steps  (98%)
[13:44:11] Completed 495000 out of 500000 steps  (99%)
[13:48:26] Completed 500000 out of 500000 steps  (100%)
[13:48:27] DynamicWrapper: Finished Work Unit: sleep=10000
[13:48:36]
[13:48:36] Finished Work Unit:
[13:48:36] - Reading up to 5030784 from "work/wudata_07.trr": Read 5030784
[13:48:36] trr file hash check passed.
[13:48:36] - Reading up to 5407360 from "work/wudata_07.xtc": Read 5407360
[13:48:37] xtc file hash check passed.
[13:48:37] edr file hash check passed.
[13:48:37] logfile size: 332901
[13:48:37] Leaving Run
[13:48:41] - Writing 10803449 bytes of core data to disk...
[13:48:43] Done: 10802937 -> 10211377 (compressed to 94.5 percent)
[13:48:43]   ... Done.
[13:48:44] - Shutting down core
[13:48:44]
[13:48:44] Folding@home Core Shutdown: FINISHED_UNIT
[13:48:48] CoreStatus = 64 (100)
[13:48:48] Unit 7 finished with 94 percent of time to deadline remaining.
[13:48:48] Updated performance fraction: 0.838189
[13:48:48] Sending work to server
[13:48:48] Project: 7504 (Run 6, Clone 223, Gen 54)


[13:48:48] + Attempting to send results [January 23 13:48:48 UTC]
[13:48:48] - Reading file work/wuresults_07.dat from core
[13:48:48]   (Read 10211889 bytes from disk)
[13:48:48] Connecting to http://128.143.199.97:8080/
[13:49:13] Posted data.
[13:49:13] Initial: 0000; - Uploaded at ~398 kB/s
[13:49:13] - Averaged speed for that direction ~317 kB/s
[13:49:13] + Results successfully sent
[13:49:13] Thank you for your contribution to Folding@Home.
[13:49:13] + Number of Units Completed: 314

[13:49:18] Trying to send all finished work units
[13:49:18] + No unsent completed units remaining.
[13:49:18] - Preparing to get new work unit...
[13:49:18] Cleaning up work directory
[13:49:18] + Attempting to get work packet
[13:49:18] Passkey found
[13:49:18] - Will indicate memory of 6135 MB
[13:49:18] - Connecting to assignment server
[13:49:18] Connecting to http://assign.stanford.edu:8080/
[13:49:18] Posted data.
[13:49:18] Initial: 8F80; - Successful: assigned to (128.143.199.97).
[13:49:18] + News From Folding@Home: Welcome to Folding@Home
[13:49:19] Loaded queue successfully.
[13:49:19] Sent data
[13:49:19] Connecting to http://128.143.199.97:8080/
[13:49:19] Posted data.
[13:49:19] Initial: 0000; - Receiving payload (expected size: 1761734)
[13:49:21] - Downloaded at ~860 kB/s
[13:49:21] - Averaged speed for that direction ~612 kB/s
[13:49:21] + Received work.
[13:49:21] Trying to send all finished work units
[13:49:21] + No unsent completed units remaining.
[13:49:21] + Closed connections
[13:49:21]
[13:49:21] + Processing work unit
[13:49:21] Core required: FahCore_a3.exe
[13:49:21] Core found.
[13:49:21] Working on queue slot 08 [January 23 13:49:21 UTC]
[13:49:21] + Working ...
[13:49:21] - Calling '.\FahCore_a3.exe -dir work/ -nice 19 -suffix 08 -np 8 -priority 96 -checkpoint 3 -service -forceasm -verbose -lifeline 4120 -version 634'

[13:49:21]
[13:49:21] *------------------------------*
[13:49:21] Folding@Home Gromacs SMP Core
[13:49:21] Version 2.27 (Dec. 15, 2010)
[13:49:21]
[13:49:21] Preparing to commence simulation
[13:49:21] - Assembly optimizations manually forced on.
[13:49:21] - Not checking prior termination.
[13:49:22] - Expanded 1761222 -> 2700832 (decompressed 153.3 percent)
[13:49:22] Called DecompressByteArray: compressed_data_size=1761222 data_size=2700832, decompressed_data_size=2700832 diff=0
[13:49:22] - Digital signature verified
[13:49:22]
[13:49:22] Project: 7504 (Run 19, Clone 106, Gen 57)
[13:49:22]
[13:49:22] Assembly optimizations on if available.
[13:49:22] Entering M.D.
[13:49:27] Mapping NT from 8 to 8
[13:49:28] Completed 0 out of 500000 steps  (0%)
[13:53:50] Completed 5000 out of 500000 steps  (1%)
[13:58:08] Completed 10000 out of 500000 steps  (2%)
[14:02:27] Completed 15000 out of 500000 steps  (3%)
[14:06:44] Completed 20000 out of 500000 steps  (4%)
[14:11:03] Completed 25000 out of 500000 steps  (5%)
[14:15:21] Completed 30000 out of 500000 steps  (6%)
[14:19:38] Completed 35000 out of 500000 steps  (7%)
[14:23:53] Completed 40000 out of 500000 steps  (8%)
[14:28:08] Completed 45000 out of 500000 steps  (9%)
[14:32:25] Completed 50000 out of 500000 steps  (10%)
[14:36:44] Completed 55000 out of 500000 steps  (11%)
[14:41:01] Completed 60000 out of 500000 steps  (12%)
[14:45:18] Completed 65000 out of 500000 steps  (13%)
[14:49:35] Completed 70000 out of 500000 steps  (14%)
[14:53:52] Completed 75000 out of 500000 steps  (15%)
[14:58:07] Completed 80000 out of 500000 steps  (16%)
[15:02:25] Completed 85000 out of 500000 steps  (17%)
[15:06:47] Completed 90000 out of 500000 steps  (18%)
[15:11:07] Completed 95000 out of 500000 steps  (19%)
[15:15:32] Completed 100000 out of 500000 steps  (20%)
[15:19:59] Completed 105000 out of 500000 steps  (21%)
[15:24:19] Completed 110000 out of 500000 steps  (22%)
[15:28:43] Completed 115000 out of 500000 steps  (23%)
[15:33:05] Completed 120000 out of 500000 steps  (24%)
[15:37:22] Completed 125000 out of 500000 steps  (25%)
[15:41:45] Completed 130000 out of 500000 steps  (26%)
[15:46:03] Completed 135000 out of 500000 steps  (27%)
[15:50:22] Completed 140000 out of 500000 steps  (28%)
[15:54:43] Completed 145000 out of 500000 steps  (29%)
[15:59:06] Completed 150000 out of 500000 steps  (30%)
[16:03:22] Completed 155000 out of 500000 steps  (31%)
[16:07:32] Completed 160000 out of 500000 steps  (32%)
[16:11:56] Completed 165000 out of 500000 steps  (33%)
[16:16:20] Completed 170000 out of 500000 steps  (34%)
[16:20:39] Completed 175000 out of 500000 steps  (35%)
[16:25:01] Completed 180000 out of 500000 steps  (36%)
[16:29:25] Completed 185000 out of 500000 steps  (37%)
[16:33:46] Completed 190000 out of 500000 steps  (38%)
[16:38:11] Completed 195000 out of 500000 steps  (39%)
[16:42:35] Completed 200000 out of 500000 steps  (40%)
[16:46:59] Completed 205000 out of 500000 steps  (41%)
[16:51:23] Completed 210000 out of 500000 steps  (42%)
[16:55:47] Completed 215000 out of 500000 steps  (43%)
[17:00:11] Completed 220000 out of 500000 steps  (44%)
[17:04:36] Completed 225000 out of 500000 steps  (45%)
[17:08:57] Completed 230000 out of 500000 steps  (46%)
[17:13:23] Completed 235000 out of 500000 steps  (47%)
[17:17:47] Completed 240000 out of 500000 steps  (48%)
[17:22:12] Completed 245000 out of 500000 steps  (49%)
[17:26:37] Completed 250000 out of 500000 steps  (50%)
[17:31:02] Completed 255000 out of 500000 steps  (51%)
[17:35:27] Completed 260000 out of 500000 steps  (52%)
[17:39:51] Completed 265000 out of 500000 steps  (53%)
[17:44:12] Completed 270000 out of 500000 steps  (54%)
[17:48:37] Completed 275000 out of 500000 steps  (55%)
[17:52:57] Completed 280000 out of 500000 steps  (56%)
[17:57:22] Completed 285000 out of 500000 steps  (57%)
[18:01:47] Completed 290000 out of 500000 steps  (58%)
[18:06:10] Completed 295000 out of 500000 steps  (59%)
[18:10:32] Completed 300000 out of 500000 steps  (60%)
[18:14:54] Completed 305000 out of 500000 steps  (61%)
[18:19:14] Completed 310000 out of 500000 steps  (62%)
[18:23:39] Completed 315000 out of 500000 steps  (63%)
[18:28:03] Completed 320000 out of 500000 steps  (64%)
[18:32:28] Completed 325000 out of 500000 steps  (65%)
[18:36:51] Completed 330000 out of 500000 steps  (66%)
[18:41:15] Completed 335000 out of 500000 steps  (67%)
[18:45:38] Completed 340000 out of 500000 steps  (68%)
[18:46:58] - Autosending finished units... [January 23 18:46:58 UTC]
[18:46:58] Trying to send all finished work units
[18:46:58] + No unsent completed units remaining.
[18:46:58] - Autosend completed
[18:50:00] Completed 345000 out of 500000 steps  (69%)
[18:54:25] Completed 350000 out of 500000 steps  (70%)
[18:58:46] Completed 355000 out of 500000 steps  (71%)
[19:03:12] Completed 360000 out of 500000 steps  (72%)
[19:07:38] Completed 365000 out of 500000 steps  (73%)
[19:12:04] Completed 370000 out of 500000 steps  (74%)
[19:16:24] Completed 375000 out of 500000 steps  (75%)
[19:20:46] Completed 380000 out of 500000 steps  (76%)
[19:25:12] Completed 385000 out of 500000 steps  (77%)
[19:29:37] Completed 390000 out of 500000 steps  (78%)
[19:33:59] Completed 395000 out of 500000 steps  (79%)
[19:38:02] Completed 400000 out of 500000 steps  (80%)
[19:42:09] Completed 405000 out of 500000 steps  (81%)
[19:46:15] Completed 410000 out of 500000 steps  (82%)
[19:50:23] Completed 415000 out of 500000 steps  (83%)
[19:54:29] Completed 420000 out of 500000 steps  (84%)
[19:58:37] Completed 425000 out of 500000 steps  (85%)
[20:02:44] Completed 430000 out of 500000 steps  (86%)
[20:06:51] Completed 435000 out of 500000 steps  (87%)
[20:10:57] Completed 440000 out of 500000 steps  (88%)
[20:15:03] Completed 445000 out of 500000 steps  (89%)
[20:19:08] Completed 450000 out of 500000 steps  (90%)
[20:23:13] Completed 455000 out of 500000 steps  (91%)
[20:27:17] Completed 460000 out of 500000 steps  (92%)
[20:31:21] Completed 465000 out of 500000 steps  (93%)
[20:35:27] Completed 470000 out of 500000 steps  (94%)
[20:39:33] Completed 475000 out of 500000 steps  (95%)
[20:43:41] Completed 480000 out of 500000 steps  (96%)
[20:47:48] Completed 485000 out of 500000 steps  (97%)
[20:51:55] Completed 490000 out of 500000 steps  (98%)
[20:56:00] Completed 495000 out of 500000 steps  (99%)
[21:00:07] Completed 500000 out of 500000 steps  (100%)
[21:00:08] DynamicWrapper: Finished Work Unit: sleep=10000
[21:00:18]
[21:00:18] Finished Work Unit:
[21:00:18] - Reading up to 5030784 from "work/wudata_08.trr": Read 5030784
[21:00:18] trr file hash check passed.
[21:00:18] - Reading up to 5405736 from "work/wudata_08.xtc": Read 5405736
[21:00:18] xtc file hash check passed.
[21:00:18] edr file hash check passed.
[21:00:18] logfile size: 333100
[21:00:18] Leaving Run
[21:00:18] - Writing 10802024 bytes of core data to disk...
[21:00:20] Done: 10801512 -> 10210749 (compressed to 94.5 percent)
[21:00:20]   ... Done.
[21:00:22] - Shutting down core
[21:00:22]
[21:00:22] Folding@home Core Shutdown: FINISHED_UNIT
[21:00:25] CoreStatus = 64 (100)
[21:00:25] Unit 8 finished with 94 percent of time to deadline remaining.
[21:00:25] Updated performance fraction: 0.859038
[21:00:25] Sending work to server
[21:00:25] Project: 7504 (Run 19, Clone 106, Gen 57)


[21:00:25] + Attempting to send results [January 23 21:00:25 UTC]
[21:00:25] - Reading file work/wuresults_08.dat from core
[21:00:25]   (Read 10211261 bytes from disk)
[21:00:25] Connecting to http://128.143.199.97:8080/
[21:00:50] Posted data.
[21:00:50] Initial: 0000; - Uploaded at ~398 kB/s
[21:00:50] - Averaged speed for that direction ~333 kB/s
[21:00:50] + Results successfully sent
[21:00:50] Thank you for your contribution to Folding@Home.
[21:00:50] + Number of Units Completed: 315

My memory is that I was getting under 4 minutes/frame in Linux for just about all WU's.

JPinTO: I'd try it were I you. Where did you read about it?
Ragnar Dan
Gerbil Elder
Silver subscriber
 
 
Posts: 5354
Joined: Sun Jan 20, 2002 7:00 pm

Re: The Hunt Is ON!

Postposted on Tue Jan 24, 2012 4:59 am

Ragnar Dan wrote:I do not have memory of any time when WinSMP was faster than Linux. I began running LinuxSMP in early 2009, and when -bigadv started it was Linux only at first IIRC, so I never ran Windows until later. The only gain from Windows was the ease of adding running a service compared to Linux.
I think it was when bigadv was screwed up in LinuxSMP (sometime late 2010/early 2011?) that WinSMP had a ppd advantage. To me the ease of adding service is a good thing as well.

Ragnar Dan wrote:I saw your post last night and decided to see how a comparison went, though the definitive way to do it would be for me to learn how to read a Windows network share from VMWare'd Linux so that I could run N frames in one system and then restart it in the other and compare results. But it appears things are more than 10 seconds per frame slower on similar WU's. Here are 2 P7504's I ran:

Code: Select all
[06:47:06] Project: 7504 (Run 6, Clone 223, Gen 54)
[06:47:06]
[06:47:06] Assembly optimizations on if available.
[06:47:06] Entering M.D.
[06:47:12] Mapping NT from 8 to 8
[06:47:13] Completed 0 out of 500000 steps  (0%)
[06:51:28] Completed 5000 out of 500000 steps  (1%)
[06:55:39] Completed 10000 out of 500000 steps  (2%)
[06:59:53] Completed 15000 out of 500000 steps  (3%)
[07:04:05] Completed 20000 out of 500000 steps  (4%)
[07:08:16] Completed 25000 out of 500000 steps  (5%)
[07:12:29] Completed 30000 out of 500000 steps  (6%)
[07:16:41] Completed 35000 out of 500000 steps  (7%)
[07:20:54] Completed 40000 out of 500000 steps  (8%)
[07:25:05] Completed 45000 out of 500000 steps  (9%)
[07:29:20] Completed 50000 out of 500000 steps  (10%)
[07:33:35] Completed 55000 out of 500000 steps  (11%)
[07:37:44] Completed 60000 out of 500000 steps  (12%)
[07:41:57] Completed 65000 out of 500000 steps  (13%)
[07:46:10] Completed 70000 out of 500000 steps  (14%)
[07:50:25] Completed 75000 out of 500000 steps  (15%)
[07:54:35] Completed 80000 out of 500000 steps  (16%)
[07:58:53] Completed 85000 out of 500000 steps  (17%)
[08:03:04] Completed 90000 out of 500000 steps  (18%)
[08:07:15] Completed 95000 out of 500000 steps  (19%)
[08:11:26] Completed 100000 out of 500000 steps  (20%)
[08:15:39] Completed 105000 out of 500000 steps  (21%)
[08:19:49] Completed 110000 out of 500000 steps  (22%)
[08:24:01] Completed 115000 out of 500000 steps  (23%)
[08:28:11] Completed 120000 out of 500000 steps  (24%)
[08:32:23] Completed 125000 out of 500000 steps  (25%)
[08:36:34] Completed 130000 out of 500000 steps  (26%)
[08:40:45] Completed 135000 out of 500000 steps  (27%)
[08:44:58] Completed 140000 out of 500000 steps  (28%)
[08:49:09] Completed 145000 out of 500000 steps  (29%)
[08:53:22] Completed 150000 out of 500000 steps  (30%)
[08:57:32] Completed 155000 out of 500000 steps  (31%)
[09:01:33] Completed 160000 out of 500000 steps  (32%)
[09:05:46] Completed 165000 out of 500000 steps  (33%)
[09:09:57] Completed 170000 out of 500000 steps  (34%)
[09:14:10] Completed 175000 out of 500000 steps  (35%)
[09:18:20] Completed 180000 out of 500000 steps  (36%)
[09:22:31] Completed 185000 out of 500000 steps  (37%)
[09:26:45] Completed 190000 out of 500000 steps  (38%)
[09:30:58] Completed 195000 out of 500000 steps  (39%)
[09:35:06] Completed 200000 out of 500000 steps  (40%)
[09:39:17] Completed 205000 out of 500000 steps  (41%)
[09:43:28] Completed 210000 out of 500000 steps  (42%)
[09:47:38] Completed 215000 out of 500000 steps  (43%)
[09:51:51] Completed 220000 out of 500000 steps  (44%)
[09:56:01] Completed 225000 out of 500000 steps  (45%)
[10:00:12] Completed 230000 out of 500000 steps  (46%)
[10:04:26] Completed 235000 out of 500000 steps  (47%)
[10:08:39] Completed 240000 out of 500000 steps  (48%)
[10:12:49] Completed 245000 out of 500000 steps  (49%)
[10:17:03] Completed 250000 out of 500000 steps  (50%)
[10:21:19] Completed 255000 out of 500000 steps  (51%)
[10:25:32] Completed 260000 out of 500000 steps  (52%)
[10:29:45] Completed 265000 out of 500000 steps  (53%)
[10:33:59] Completed 270000 out of 500000 steps  (54%)
[10:38:13] Completed 275000 out of 500000 steps  (55%)
[10:42:26] Completed 280000 out of 500000 steps  (56%)
[10:46:39] Completed 285000 out of 500000 steps  (57%)
[10:50:53] Completed 290000 out of 500000 steps  (58%)
[10:55:05] Completed 295000 out of 500000 steps  (59%)
[10:59:17] Completed 300000 out of 500000 steps  (60%)
[11:03:30] Completed 305000 out of 500000 steps  (61%)
[11:07:42] Completed 310000 out of 500000 steps  (62%)
[11:11:57] Completed 315000 out of 500000 steps  (63%)
[11:16:08] Completed 320000 out of 500000 steps  (64%)
[11:20:21] Completed 325000 out of 500000 steps  (65%)
[11:24:32] Completed 330000 out of 500000 steps  (66%)
[11:28:46] Completed 335000 out of 500000 steps  (67%)
[11:33:01] Completed 340000 out of 500000 steps  (68%)
[11:37:11] Completed 345000 out of 500000 steps  (69%)
[11:41:27] Completed 350000 out of 500000 steps  (70%)
[11:45:39] Completed 355000 out of 500000 steps  (71%)
[11:49:54] Completed 360000 out of 500000 steps  (72%)
[11:54:05] Completed 365000 out of 500000 steps  (73%)
[11:58:17] Completed 370000 out of 500000 steps  (74%)
[12:02:30] Completed 375000 out of 500000 steps  (75%)
[12:06:43] Completed 380000 out of 500000 steps  (76%)
[12:10:56] Completed 385000 out of 500000 steps  (77%)
[12:15:08] Completed 390000 out of 500000 steps  (78%)
[12:19:19] Completed 395000 out of 500000 steps  (79%)
[12:23:31] Completed 400000 out of 500000 steps  (80%)
[12:27:42] Completed 405000 out of 500000 steps  (81%)
[12:31:52] Completed 410000 out of 500000 steps  (82%)
[12:36:00] Completed 415000 out of 500000 steps  (83%)
[12:40:17] Completed 420000 out of 500000 steps  (84%)
[12:44:32] Completed 425000 out of 500000 steps  (85%)
[12:46:58] - Autosending finished units... [January 23 12:46:58 UTC]
[12:46:58] Trying to send all finished work units
[12:46:58] + No unsent completed units remaining.
[12:46:58] - Autosend completed
[12:48:47] Completed 430000 out of 500000 steps  (86%)
[12:53:03] Completed 435000 out of 500000 steps  (87%)
[12:57:16] Completed 440000 out of 500000 steps  (88%)
[13:01:31] Completed 445000 out of 500000 steps  (89%)
[13:05:47] Completed 450000 out of 500000 steps  (90%)
[13:10:02] Completed 455000 out of 500000 steps  (91%)
[13:14:19] Completed 460000 out of 500000 steps  (92%)
[13:18:34] Completed 465000 out of 500000 steps  (93%)
[13:22:51] Completed 470000 out of 500000 steps  (94%)
[13:27:11] Completed 475000 out of 500000 steps  (95%)
[13:31:26] Completed 480000 out of 500000 steps  (96%)
[13:35:41] Completed 485000 out of 500000 steps  (97%)
[13:39:57] Completed 490000 out of 500000 steps  (98%)
[13:44:11] Completed 495000 out of 500000 steps  (99%)
[13:48:26] Completed 500000 out of 500000 steps  (100%)
[13:48:27] DynamicWrapper: Finished Work Unit: sleep=10000
[13:48:36]
[13:48:36] Finished Work Unit:
[13:48:36] - Reading up to 5030784 from "work/wudata_07.trr": Read 5030784
[13:48:36] trr file hash check passed.
[13:48:36] - Reading up to 5407360 from "work/wudata_07.xtc": Read 5407360
[13:48:37] xtc file hash check passed.
[13:48:37] edr file hash check passed.
[13:48:37] logfile size: 332901
[13:48:37] Leaving Run
[13:48:41] - Writing 10803449 bytes of core data to disk...
[13:48:43] Done: 10802937 -> 10211377 (compressed to 94.5 percent)
[13:48:43]   ... Done.
[13:48:44] - Shutting down core
[13:48:44]
[13:48:44] Folding@home Core Shutdown: FINISHED_UNIT
[13:48:48] CoreStatus = 64 (100)
[13:48:48] Unit 7 finished with 94 percent of time to deadline remaining.
[13:48:48] Updated performance fraction: 0.838189
[13:48:48] Sending work to server
[13:48:48] Project: 7504 (Run 6, Clone 223, Gen 54)


[13:48:48] + Attempting to send results [January 23 13:48:48 UTC]
[13:48:48] - Reading file work/wuresults_07.dat from core
[13:48:48]   (Read 10211889 bytes from disk)
[13:48:48] Connecting to http://128.143.199.97:8080/
[13:49:13] Posted data.
[13:49:13] Initial: 0000; - Uploaded at ~398 kB/s
[13:49:13] - Averaged speed for that direction ~317 kB/s
[13:49:13] + Results successfully sent
[13:49:13] Thank you for your contribution to Folding@Home.
[13:49:13] + Number of Units Completed: 314

[13:49:18] Trying to send all finished work units
[13:49:18] + No unsent completed units remaining.
[13:49:18] - Preparing to get new work unit...
[13:49:18] Cleaning up work directory
[13:49:18] + Attempting to get work packet
[13:49:18] Passkey found
[13:49:18] - Will indicate memory of 6135 MB
[13:49:18] - Connecting to assignment server
[13:49:18] Connecting to http://assign.stanford.edu:8080/
[13:49:18] Posted data.
[13:49:18] Initial: 8F80; - Successful: assigned to (128.143.199.97).
[13:49:18] + News From Folding@Home: Welcome to Folding@Home
[13:49:19] Loaded queue successfully.
[13:49:19] Sent data
[13:49:19] Connecting to http://128.143.199.97:8080/
[13:49:19] Posted data.
[13:49:19] Initial: 0000; - Receiving payload (expected size: 1761734)
[13:49:21] - Downloaded at ~860 kB/s
[13:49:21] - Averaged speed for that direction ~612 kB/s
[13:49:21] + Received work.
[13:49:21] Trying to send all finished work units
[13:49:21] + No unsent completed units remaining.
[13:49:21] + Closed connections
[13:49:21]
[13:49:21] + Processing work unit
[13:49:21] Core required: FahCore_a3.exe
[13:49:21] Core found.
[13:49:21] Working on queue slot 08 [January 23 13:49:21 UTC]
[13:49:21] + Working ...
[13:49:21] - Calling '.\FahCore_a3.exe -dir work/ -nice 19 -suffix 08 -np 8 -priority 96 -checkpoint 3 -service -forceasm -verbose -lifeline 4120 -version 634'

[13:49:21]
[13:49:21] *------------------------------*
[13:49:21] Folding@Home Gromacs SMP Core
[13:49:21] Version 2.27 (Dec. 15, 2010)
[13:49:21]
[13:49:21] Preparing to commence simulation
[13:49:21] - Assembly optimizations manually forced on.
[13:49:21] - Not checking prior termination.
[13:49:22] - Expanded 1761222 -> 2700832 (decompressed 153.3 percent)
[13:49:22] Called DecompressByteArray: compressed_data_size=1761222 data_size=2700832, decompressed_data_size=2700832 diff=0
[13:49:22] - Digital signature verified
[13:49:22]
[13:49:22] Project: 7504 (Run 19, Clone 106, Gen 57)
[13:49:22]
[13:49:22] Assembly optimizations on if available.
[13:49:22] Entering M.D.
[13:49:27] Mapping NT from 8 to 8
[13:49:28] Completed 0 out of 500000 steps  (0%)
[13:53:50] Completed 5000 out of 500000 steps  (1%)
[13:58:08] Completed 10000 out of 500000 steps  (2%)
[14:02:27] Completed 15000 out of 500000 steps  (3%)
[14:06:44] Completed 20000 out of 500000 steps  (4%)
[14:11:03] Completed 25000 out of 500000 steps  (5%)
[14:15:21] Completed 30000 out of 500000 steps  (6%)
[14:19:38] Completed 35000 out of 500000 steps  (7%)
[14:23:53] Completed 40000 out of 500000 steps  (8%)
[14:28:08] Completed 45000 out of 500000 steps  (9%)
[14:32:25] Completed 50000 out of 500000 steps  (10%)
[14:36:44] Completed 55000 out of 500000 steps  (11%)
[14:41:01] Completed 60000 out of 500000 steps  (12%)
[14:45:18] Completed 65000 out of 500000 steps  (13%)
[14:49:35] Completed 70000 out of 500000 steps  (14%)
[14:53:52] Completed 75000 out of 500000 steps  (15%)
[14:58:07] Completed 80000 out of 500000 steps  (16%)
[15:02:25] Completed 85000 out of 500000 steps  (17%)
[15:06:47] Completed 90000 out of 500000 steps  (18%)
[15:11:07] Completed 95000 out of 500000 steps  (19%)
[15:15:32] Completed 100000 out of 500000 steps  (20%)
[15:19:59] Completed 105000 out of 500000 steps  (21%)
[15:24:19] Completed 110000 out of 500000 steps  (22%)
[15:28:43] Completed 115000 out of 500000 steps  (23%)
[15:33:05] Completed 120000 out of 500000 steps  (24%)
[15:37:22] Completed 125000 out of 500000 steps  (25%)
[15:41:45] Completed 130000 out of 500000 steps  (26%)
[15:46:03] Completed 135000 out of 500000 steps  (27%)
[15:50:22] Completed 140000 out of 500000 steps  (28%)
[15:54:43] Completed 145000 out of 500000 steps  (29%)
[15:59:06] Completed 150000 out of 500000 steps  (30%)
[16:03:22] Completed 155000 out of 500000 steps  (31%)
[16:07:32] Completed 160000 out of 500000 steps  (32%)
[16:11:56] Completed 165000 out of 500000 steps  (33%)
[16:16:20] Completed 170000 out of 500000 steps  (34%)
[16:20:39] Completed 175000 out of 500000 steps  (35%)
[16:25:01] Completed 180000 out of 500000 steps  (36%)
[16:29:25] Completed 185000 out of 500000 steps  (37%)
[16:33:46] Completed 190000 out of 500000 steps  (38%)
[16:38:11] Completed 195000 out of 500000 steps  (39%)
[16:42:35] Completed 200000 out of 500000 steps  (40%)
[16:46:59] Completed 205000 out of 500000 steps  (41%)
[16:51:23] Completed 210000 out of 500000 steps  (42%)
[16:55:47] Completed 215000 out of 500000 steps  (43%)
[17:00:11] Completed 220000 out of 500000 steps  (44%)
[17:04:36] Completed 225000 out of 500000 steps  (45%)
[17:08:57] Completed 230000 out of 500000 steps  (46%)
[17:13:23] Completed 235000 out of 500000 steps  (47%)
[17:17:47] Completed 240000 out of 500000 steps  (48%)
[17:22:12] Completed 245000 out of 500000 steps  (49%)
[17:26:37] Completed 250000 out of 500000 steps  (50%)
[17:31:02] Completed 255000 out of 500000 steps  (51%)
[17:35:27] Completed 260000 out of 500000 steps  (52%)
[17:39:51] Completed 265000 out of 500000 steps  (53%)
[17:44:12] Completed 270000 out of 500000 steps  (54%)
[17:48:37] Completed 275000 out of 500000 steps  (55%)
[17:52:57] Completed 280000 out of 500000 steps  (56%)
[17:57:22] Completed 285000 out of 500000 steps  (57%)
[18:01:47] Completed 290000 out of 500000 steps  (58%)
[18:06:10] Completed 295000 out of 500000 steps  (59%)
[18:10:32] Completed 300000 out of 500000 steps  (60%)
[18:14:54] Completed 305000 out of 500000 steps  (61%)
[18:19:14] Completed 310000 out of 500000 steps  (62%)
[18:23:39] Completed 315000 out of 500000 steps  (63%)
[18:28:03] Completed 320000 out of 500000 steps  (64%)
[18:32:28] Completed 325000 out of 500000 steps  (65%)
[18:36:51] Completed 330000 out of 500000 steps  (66%)
[18:41:15] Completed 335000 out of 500000 steps  (67%)
[18:45:38] Completed 340000 out of 500000 steps  (68%)
[18:46:58] - Autosending finished units... [January 23 18:46:58 UTC]
[18:46:58] Trying to send all finished work units
[18:46:58] + No unsent completed units remaining.
[18:46:58] - Autosend completed
[18:50:00] Completed 345000 out of 500000 steps  (69%)
[18:54:25] Completed 350000 out of 500000 steps  (70%)
[18:58:46] Completed 355000 out of 500000 steps  (71%)
[19:03:12] Completed 360000 out of 500000 steps  (72%)
[19:07:38] Completed 365000 out of 500000 steps  (73%)
[19:12:04] Completed 370000 out of 500000 steps  (74%)
[19:16:24] Completed 375000 out of 500000 steps  (75%)
[19:20:46] Completed 380000 out of 500000 steps  (76%)
[19:25:12] Completed 385000 out of 500000 steps  (77%)
[19:29:37] Completed 390000 out of 500000 steps  (78%)
[19:33:59] Completed 395000 out of 500000 steps  (79%)
[19:38:02] Completed 400000 out of 500000 steps  (80%)
[19:42:09] Completed 405000 out of 500000 steps  (81%)
[19:46:15] Completed 410000 out of 500000 steps  (82%)
[19:50:23] Completed 415000 out of 500000 steps  (83%)
[19:54:29] Completed 420000 out of 500000 steps  (84%)
[19:58:37] Completed 425000 out of 500000 steps  (85%)
[20:02:44] Completed 430000 out of 500000 steps  (86%)
[20:06:51] Completed 435000 out of 500000 steps  (87%)
[20:10:57] Completed 440000 out of 500000 steps  (88%)
[20:15:03] Completed 445000 out of 500000 steps  (89%)
[20:19:08] Completed 450000 out of 500000 steps  (90%)
[20:23:13] Completed 455000 out of 500000 steps  (91%)
[20:27:17] Completed 460000 out of 500000 steps  (92%)
[20:31:21] Completed 465000 out of 500000 steps  (93%)
[20:35:27] Completed 470000 out of 500000 steps  (94%)
[20:39:33] Completed 475000 out of 500000 steps  (95%)
[20:43:41] Completed 480000 out of 500000 steps  (96%)
[20:47:48] Completed 485000 out of 500000 steps  (97%)
[20:51:55] Completed 490000 out of 500000 steps  (98%)
[20:56:00] Completed 495000 out of 500000 steps  (99%)
[21:00:07] Completed 500000 out of 500000 steps  (100%)
[21:00:08] DynamicWrapper: Finished Work Unit: sleep=10000
[21:00:18]
[21:00:18] Finished Work Unit:
[21:00:18] - Reading up to 5030784 from "work/wudata_08.trr": Read 5030784
[21:00:18] trr file hash check passed.
[21:00:18] - Reading up to 5405736 from "work/wudata_08.xtc": Read 5405736
[21:00:18] xtc file hash check passed.
[21:00:18] edr file hash check passed.
[21:00:18] logfile size: 333100
[21:00:18] Leaving Run
[21:00:18] - Writing 10802024 bytes of core data to disk...
[21:00:20] Done: 10801512 -> 10210749 (compressed to 94.5 percent)
[21:00:20]   ... Done.
[21:00:22] - Shutting down core
[21:00:22]
[21:00:22] Folding@home Core Shutdown: FINISHED_UNIT
[21:00:25] CoreStatus = 64 (100)
[21:00:25] Unit 8 finished with 94 percent of time to deadline remaining.
[21:00:25] Updated performance fraction: 0.859038
[21:00:25] Sending work to server
[21:00:25] Project: 7504 (Run 19, Clone 106, Gen 57)


[21:00:25] + Attempting to send results [January 23 21:00:25 UTC]
[21:00:25] - Reading file work/wuresults_08.dat from core
[21:00:25]   (Read 10211261 bytes from disk)
[21:00:25] Connecting to http://128.143.199.97:8080/
[21:00:50] Posted data.
[21:00:50] Initial: 0000; - Uploaded at ~398 kB/s
[21:00:50] - Averaged speed for that direction ~333 kB/s
[21:00:50] + Results successfully sent
[21:00:50] Thank you for your contribution to Folding@Home.
[21:00:50] + Number of Units Completed: 315

My memory is that I was getting under 4 minutes/frame in Linux for just about all WU's.

My log file only contains 1 sample of P7504. Looks like 6 minutes / frame. But I am running -smp 7, and on an i875K @2.93GHz stock, with regular use in the foreground.
Code: Select all
[11:21:52] Project: 7504 (Run 10, Clone 18, Gen 37)
[11:21:52]
[11:21:52] Assembly optimizations on if available.
[11:21:52] Entering M.D.
[11:21:58] Mapping NT from 7 to 7
[11:21:59] Completed 0 out of 500000 steps  (0%)
[11:27:48] Completed 5000 out of 500000 steps  (1%)
[11:33:39] Completed 10000 out of 500000 steps  (2%)
[11:39:29] Completed 15000 out of 500000 steps  (3%)
[11:45:19] Completed 20000 out of 500000 steps  (4%)
[11:51:10] Completed 25000 out of 500000 steps  (5%)
[11:57:00] Completed 30000 out of 500000 steps  (6%)
[12:02:51] Completed 35000 out of 500000 steps  (7%)
[12:08:42] Completed 40000 out of 500000 steps  (8%)
[12:14:32] Completed 45000 out of 500000 steps  (9%)
[12:20:24] Completed 50000 out of 500000 steps  (10%)
[12:26:15] Completed 55000 out of 500000 steps  (11%)
[12:32:06] Completed 60000 out of 500000 steps  (12%)
[12:37:57] Completed 65000 out of 500000 steps  (13%)
[12:43:47] Completed 70000 out of 500000 steps  (14%)
[12:49:39] Completed 75000 out of 500000 steps  (15%)
[12:55:31] Completed 80000 out of 500000 steps  (16%)
[13:01:23] Completed 85000 out of 500000 steps  (17%)


Ragnar Dan wrote:JPinTO: I'd try it were I you. Where did you read about it?
What's this core count spoofing thing? :o
Image
The Model M is not for the faint of heart. You either like them or hate them.

Gerbils unite! Fold for UnitedGerbilNation, team 2630.
Flying Fox
Gerbil God
 
Posts: 24289
Joined: Mon May 24, 2004 2:19 am

Re: The Hunt Is ON!

Postposted on Thu Jan 26, 2012 8:35 pm

Flying Fox wrote:
Ragnar Dan wrote:JPinTO: I'd try it were I you. Where did you read about it?
What's this core count spoofing thing? :o

The questions seem to have stumped our interlocutor. :wink:
Ragnar Dan
Gerbil Elder
Silver subscriber
 
 
Posts: 5354
Joined: Sun Jan 20, 2002 7:00 pm

Re: The Hunt Is ON!

Postposted on Sat Jan 28, 2012 8:24 pm

Spoofing linux to report whatever number of cores you wish it to:

http://www.overclock.net/t/1048365/ubun ... ated-2600k
JPinTO
Gerbil Team Leader
 
Posts: 240
Joined: Sat Jun 30, 2007 7:02 am
Location: Toronto, Ontario

Re: The Hunt Is ON!

Postposted on Mon Feb 06, 2012 11:10 am

That's interesting looking. I wonder why what of it I've seen describes such a large install and wants you to burn a disc with the Ubuntu ISO, but I haven't read the whole thing yet. Somewhere on the folding forum I found a folding-specific Linux VM for VMWare which was fairly small and I'd been using that. But it has its own problems such as the filesystem it uses, and the last time I tried to run it, it seemed to be screwed up and kept crashing the folding client or some such thing (I forget exactly... I should write this stuff down, but having quit most of my folding I didn't consider it particularly important), though it worked well for quite a while before whatever I must've done to screw it up.

Anyway, someone with Linux and VM knowledge should come up with a combination of these things and make the disk image smaller and more specific.
Ragnar Dan
Gerbil Elder
Silver subscriber
 
 
Posts: 5354
Joined: Sun Jan 20, 2002 7:00 pm

Re: The Hunt Is ON!

Postposted on Sun Feb 19, 2012 7:30 am

Hey Fuzzhead... Do you mind giving me some info on the farm required to generate 300k PPD?

I'm barely getting 50k PPD with 3x SMP clients: i7-920, i7-2600K and 970X. I'm doing something wrong.

- JP
JPinTO
Gerbil Team Leader
 
Posts: 240
Joined: Sat Jun 30, 2007 7:02 am
Location: Toronto, Ontario

Re: The Hunt Is ON!

Postposted on Fri Mar 02, 2012 2:57 pm

Overclock the hell out them! My 920 and 930 are at 4.1 GHz. You have more head room.
The 3 970's are at 4.2Ghz. My 2700k is at 4.7Ghz. I only have two GPU's folding it the moment a 560Ti and a 450 both factory overclocked. The other 450 is RMAed do to fan failure.
You can still do big adv on the 970. You can run Linux like I do on two of them Ubuntu 11.10 and SUSE 12.1 on the other. Or like on my main box I run Oracle Virtualbox and Ubuntu 10.4.3. Vitualbox is better than VMware as it allows more cores/threads. And do not run a gpu
at the same time or you will not meet the deadlines. Stanford is still giving out BA12's that's 12 threads on Linux. Not sure how long it will last.
As far as the Linux core count trick, that is why Stanford is changing the Big Adv policy.
So I do not condone the trick/cheating

HFM shows 313,554 ppd at this time.

Fuzzhead :D
fuzzhead
Gerbil First Class
Gold subscriber
 
 
Posts: 175
Joined: Thu Mar 10, 2005 12:02 am
Location: Florida

Re: The Hunt Is ON!

Postposted on Fri Jun 01, 2012 5:24 pm

I added another system to push myself into the top 10 for 2630.
Breaking into the top 5 looks bleak unless I add 3 more machines.
I dont think this one will help much http://www.newegg.com/Product/Product.a ... 6819105017
It sounds like a jet taking off when I boot it up.
ASUS P5B-E,ConroeE6400,2GB Mushkin DDR2 800,
EVGA 8800GTS,Corsair 520HX,Antec 900,WD320 GB,Samsung 204B
rogue426
Graphmaster Gerbil
 
Posts: 1251
Joined: Thu Jul 21, 2005 9:51 pm

Previous

Return to TR Distributed Computing Effort

Who is online

Users browsing this forum: No registered users and 2 guests