| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Beta Testing
Forum: Beta Test Support Forum
Thread: Discovering Dengue Drugs - Together Phase 2 BETA |
| No member browsing this thread |
|
Thread Status: Locked Total posts in this thread: 192
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
After perusing the bandwidth requirements in
http://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,27642#252600 I was surprised to see my Fedora quads (4GB Reaper DDR2 each) picked up 6 typeA WUs between them when the fastest download speed I've ever gotten out of my Novatel 3G card here at home (4 miles from nearest tower) is about 160Kbps, and 60Kbps up. No windows machines here got any beta WUs so far, though... so maybe that was an exception made for Linux platforms? Anyway... It took about 1 hr 40 mins for the 2.4GHz Intel quad to checkpoint; 1 hr 20 mins to the first checkpoint on the AMD 3GHz quad... initial estimate was about ~48 hours on all, and the 2.4GHz Intel appears to be sticking to that; but after 1 checkpoint the 3GHz AMD has decided it should finish them in about 34 hours. Host Project Date Message Hope that info helps.  |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Someone asked, think someone dug into the code and it reminded of the same when ACAH-1 ran. BOINC computes with each upload/download what the effective rate of transfer is. If your prefs / default device profile settings are set to zero, BOINC will always try to use the maximum. By past comments there are some versions around where the setting apparently has no effect, but if you run 6.2.28 and up you should be okay if you wish to control the BOINC used bandwidth. Mine are set to 512 Mb, both ways.
----------------------------------------When you watch the transfer window in BOINC, you see rates of uncompressed data. Everything is dynamically compressed during transfer. HPF2 is pre-compressed already down/uploading. Think that's old but still valid info.
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
widdershins
Veteran Cruncher Scotland Joined: Apr 30, 2007 Post Count: 674 Status: Offline Project Badges: 
|
I was a bit concerned that my Quad would run out of memory when 4 of these were going at the same time as it only has 4Gb of Ram so I tweaked the qeue to start four together whilst I was still awake and then ran "top" to see the true memory usage on my Linux box.
It seems as though the key factor is more the swapfile than the "real"memory. The jobs seem to be running with a 1.5G swap file each. Page fault count is tiny considering how much they're using.  |
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges: 
|
yes, the virtual memory is the key on these work units. Also, CPU time is huge... We are working on decreasing the VM but it probably won't make it by project launch. But it is a goal to decrease these requirements.
-Uplinger |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
hmmm, reminds me of the Linux issue with suspected swapfile exceeding and BOINC not keeping track. Better make folk aware to set VM to auto-sizing with ample space to grow. An I920 with 8 concurrently will cause some stress :P
----------------------------------------
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
David_L6
Senior Cruncher USA Joined: Aug 24, 2006 Post Count: 296 Status: Offline Project Badges: 
|
Four errors so far. Two on a machine that is overclocked pretty good (Vista Ultimate 64 bit) and just may throw an error or two once in a while. Two on a machine (XP Pro 32 bit) that is overclocked slightly and hardly ever throws an error.
----------------------------------------Vista Ultimate 64 bit, Q6700, 8GB RAM: One pending validation (me - BETA_erag_a057_ps0005 - 3.45 hours). This one has one error and two in progress. 1st error: just me so far, 3 in progress. - exit code -1073741819 (0xc0000005) 2nd error: 7 errors, 1 "other". The system cannot write to the specified device. (0x1d) - exit code 29 (0x1d) XP Pro 32 bit, QX6700, 4GB RAM: 1st error: mine errored, one User Aborted, two In Progress, one Waiting to be sent. The system cannot write to the specified device. (0x1d) - exit code 29 (0x1d) 2nd error: 4 errors, one In Progress, two Waiting to be sent. The system cannot write to the specified device. (0x1d) - exit code 29 (0x1d) I hope that this info helps. I know the risks of running Betas and accept that risk to help the projects in the long run. I am NOT complaining - just reporting!  [Edit 1 times, last edit by David_L6 at Oct 7, 2009 1:44:14 AM] |
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges:
|
David, thanks for the report. The exit code 29 is something in the work unit that we are going to be working on. Seems your machines got unlucky with 3 of them. the other one is a memory access issue, could be many things...
Thanks, -Uplinger |
||
|
|
TimAndHedy
Senior Cruncher Joined: Jan 27, 2009 Post Count: 267 Status: Offline Project Badges: 
|
I got home and noticed a status "Computation Error" in the BOINC Client
----------------------------------------for BETA_ erag_ a006_ ps0000_ 3 I was not sure it this was dealt with yet or not but here is the log output. This has not been uploaded yet since the time to completion for the 3 other betas is 74 hours. I think the 74 hour thing is related to a very slow HFCC unit that took 160% longer than normal. 06-Oct-2009 18:16:23 [World Community Grid] Starting BETA_erag_a006_ps0000_3 06-Oct-2009 18:16:30 [World Community Grid] Starting task BETA_erag_a006_ps0000_3 using beta10 version 607 06-Oct-2009 18:16:31 [World Community Grid] Started upload of HFCC_t1_02464821_TrkB_0000_0_0 06-Oct-2009 18:16:31 [World Community Grid] Started upload of HFCC_t1_02464821_TrkB_0000_0_1 06-Oct-2009 18:16:34 [World Community Grid] Finished upload of HFCC_t1_02464821_TrkB_0000_0_1 06-Oct-2009 18:16:34 [World Community Grid] Started upload of HFCC_t1_02464821_TrkB_0000_0_2 06-Oct-2009 18:16:36 [World Community Grid] Finished upload of HFCC_t1_02464821_TrkB_0000_0_0 06-Oct-2009 18:16:36 [World Community Grid] Finished upload of HFCC_t1_02464821_TrkB_0000_0_2 06-Oct-2009 18:16:36 [World Community Grid] Started upload of HFCC_t1_02464821_TrkB_0000_0_3 06-Oct-2009 18:16:38 [World Community Grid] Finished upload of HFCC_t1_02464821_TrkB_0000_0_3 06-Oct-2009 18:25:32 [World Community Grid] Computation for task BETA_erag_a006_ps0000_3 finished 06-Oct-2009 18:25:33 [World Community Grid] Output file BETA_erag_a006_ps0000_3_4 for task BETA_erag_a006_ps0000_3 absent 06-Oct-2009 18:25:33 [World Community Grid] Output file BETA_erag_a006_ps0000_3_5 for task BETA_erag_a006_ps0000_3 absent 06-Oct-2009 18:25:33 [World Community Grid] Output file BETA_erag_a006_ps0000_3_6 for task BETA_erag_a006_ps0000_3 absent 06-Oct-2009 18:25:33 [World Community Grid] Starting X0000084020991200703091944_0 06-Oct-2009 18:25:33 [World Community Grid] Starting task X0000084020991200703091944_0 using hcc1 version 606 06-Oct-2009 18:25:34 [World Community Grid] Started upload of BETA_erag_a006_ps0000_3_0 06-Oct-2009 18:25:34 [World Community Grid] Started upload of BETA_erag_a006_ps0000_3_1 06-Oct-2009 18:25:36 [World Community Grid] Finished upload of BETA_erag_a006_ps0000_3_1 06-Oct-2009 18:25:36 [World Community Grid] Started upload of BETA_erag_a006_ps0000_3_2 06-Oct-2009 18:25:39 [World Community Grid] Finished upload of BETA_erag_a006_ps0000_3_0 06-Oct-2009 18:25:39 [World Community Grid] Started upload of BETA_erag_a006_ps0000_3_3 06-Oct-2009 18:25:42 [World Community Grid] Finished upload of BETA_erag_a006_ps0000_3_3 06-Oct-2009 18:28:46 [World Community Grid] Finished upload of BETA_erag_a006_ps0000_3_2 [Edit 2 times, last edit by TimAndHedy at Oct 7, 2009 2:00:24 AM] |
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges:
|
TimAndHedy, This is probably and error 29 that I was describing before, what you see is that 3 files that are expected did not get written due to this exit code. Basically in this case it means the work unit did not finish completely and the expected files that were required to be uploaded did not exist. If you check your result status page on the website you will probably see something refering to exit 29.
-Uplinger |
||
|
|
X-Files 27
Senior Cruncher Canada Joined: May 21, 2007 Post Count: 391 Status: Offline Project Badges:
|
got only 2 beta units...1 is done and errored out with exit code 29
----------------------------------------the other WU is 7.3% Done at 3hrs with 38hrs remaining on q9450 stock speed.   [Edit 1 times, last edit by X-Files 27 at Oct 7, 2009 6:57:59 AM] |
||
|
|
|