charmm_5.4_windows_amd64
Message boards : Application Info : charmm_5.4_windows_amd64
| Author | Message | |
|---|---|---|
Andre Kerstens
Forum moderator Project tester Volunteer tester 
Joined: Sep 11 06 Posts: 749 ID: 1 Credit: 15,199 RAC: 0 |
||
|
This version can be picked up by the 64-bit BOINC client and run on 64-bit machines. It is basically a wrapper around the 32-bit version of Charmm, so do not expect any performance gains from this binary.
|
||
ID:
2587 | Rating: 0
| rate:
/
|
||
|
This version can be picked up by the 64-bit BOINC client and run on 64-bit machines. It is basically a wrapper around the 32-bit version of Charmm, so do not expect any performance gains from this binary. Is there a link to a preferred build of the Windows 64-bit BOINC core client, or are we free to use any? - I'll most likely be using Crunch3r's. |
||
|
ID:
2741 | Rating: 0
| rate:
/
|
||
|
Andre Kerstens
Forum moderator Project tester Volunteer tester
Joined: Sep 11 06 Posts: 749 ID: 1 Credit: 15,199 RAC: 0 |
||
|
Please feel free to use any, although Must say that I don't have much experience with the 64-bit clients. At least it doesn't break our app and that's good enough :-)
This version can be picked up by the 64-bit BOINC client and run on 64-bit machines. It is basically a wrapper around the 32-bit version of Charmm, so do not expect any performance gains from this binary. ____________ D@H the greatest project in the world... a while from now! |
||
|
ID:
2742 | Rating: 0
| rate:
/
|
||
|
Please feel free to use any, although Must say that I don't have much experience with the 64-bit clients. At least it doesn't break our app and that's good enough :-) It's going to get even more monstrous.. lol as soon as I get this all working I will place my order for the QX6700 (Quad Core) I just want to be sure 64-Bit Vista is all that it's been hyped so this is really just a test machine for me now. No use dropping 1k on a processor until I have it mostly working and I'm glad I didn't do that 2 months ago when I ordered all this stuff as I had a very depressing experience practicing with Windows XP 64-bit. The *official* 64-bit Vista install (on this machine) will begin when I get my new RAID array so that I can backup the old one onto it. (It's in the mail? Somewhere? On it's way! I can't wait! Be here anyday!) Until then I'll play using my Windows XP 64-Bit install, but that pretty Vista CD is making me pretty itchy. I'm moving from 8x300gb ATA/133 Maxtor Diamondmax 10's (currently 4 RAID0 arrays) to 4(with spare)x750GB SATA Seagate Barracuda ES's. So this will also be my first experiment with RAID5. I'm guessing I won't be losing much room we'll see how it all comes out. They aren't completely full yet anyway so I have a little wiggle room. P.S. It's watercooled, but I don't overclock.. ha! isn't that sad.... running a full dual processor load on it all day and the CPU temp is 3 degrees lower than room temperature... lololol Very pretty machine tho. Right now this is what I got.. Thermaltake Kandalf LCS Liquid Cooled Tower Thermaltake 1200 Watt Power Supply Intel Core 2 Duo X6700 ASUS P5W DH Deluxe Motherboard 4 sticks of 1GB 3-3-3-8 DDR2-800Mhz GEIL Memory Adaptec SCSI 29160N SCSI Adapter Seagate 15k.5 SCSI drive (15,000 RPM) Dual ATI X1950 XTX Crossfire Cards Creative Labs Sound Blaster X-Fi Elite Pro (more later..) |
||
|
ID:
2745 | Rating: 0
| rate:
/
|
||
|
Finally ran the cache out of my 32-bit BoincMGR, and the first set of 15 32-bit test WU are crunching along peacefully on the 32 bit BOINC.
|
||
|
ID:
2801 | Rating: 0
| rate:
/
|
||
|
3/26/2007 11:39:55 AM||General prefs: using your defaults
|
||
|
ID:
2817 | Rating: 0
| rate:
/
|
||
|
3/26/2007 11:39:55 AM||General prefs: using your defaults There isn't any "true" 64-bit WUs for D@H. What we get is 32-bit WUs in a wrapper, so the 64-bit Linux will run them. Right now there isn't any of those to be had either. I just suspended D@H until they have the new WUs ready. I don't like BOINC adding time to long-term debt, while we don't get WUs. I want to have the choice as to how much computer D@H gets when WUs start again, not have the scheduler give all to D@H until share is balanced. |
||
|
ID:
2819 | Rating: 0
| rate:
/
|
||
|
There isn't any "true" 64-bit WUs for D@H. What we get is 32-bit WUs in a wrapper, so the 64-bit Linux will run them. Right now there isn't any of those to be had either. I just suspended D@H until they have the new WUs ready. I don't like BOINC adding time to long-term debt, while we don't get WUs. I want to have the choice as to how much computer D@H gets when WUs start again, not have the scheduler give all to D@H until share is balanced. Well, yes.. I was just saying that I'm not getting any that were committed to my platform. (windows_amd64) Of course.. if there are none being sent out for Linux64, WIndows32bit, or Windows64bit... Why are there 400 wu's waiting to be sent? Are they *all* committed to 32-bit linux and Apple-Darwin? Come to think of it.. that number's been at 407 for quite a while.. |
||
|
ID:
2820 | Rating: 0
| rate:
/
|
||
|
There isn't any "true" 64-bit WUs for D@H. What we get is 32-bit WUs in a wrapper, so the 64-bit Linux will run them. Right now there isn't any of those to be had either. I just suspended D@H until they have the new WUs ready. I don't like BOINC adding time to long-term debt, while we don't get WUs. I want to have the choice as to how much computer D@H gets when WUs start again, not have the scheduler give all to D@H until share is balanced. Yep, 362 for MAC Intel for several days. |
||
|
ID:
2822 | Rating: 0
| rate:
/
|
||
|
David Ball
Forum moderator Volunteer tester 
Joined: Sep 18 06 Posts: 274 ID: 115 Credit: 1,634,401 RAC: 0 |
||
|
I believe they found some problems that only showed up when the work queues were drained. IIRC, Andre said that those were very old work units that the system is trying to send out again in order to meet a quorum. My guess would be that getting everything else out of the way is now letting them find the work units that were orphaned by earlier changes in the HR and Quorum rules plus changes in the BOINC client software. The new BOINC clients return different cpu id information and those old WUs probably can't find anything to match.
|
||
|
ID:
2824 | Rating: 0
| rate:
/
|
||
|
I believe they found some problems that only showed up when the work queues were drained. IIRC, Andre said that those were very old work units that the system is trying to send out again in order to meet a quorum. My guess would be that getting everything else out of the way is now letting them find the work units that were orphaned by earlier changes in the HR and Quorum rules plus changes in the BOINC client software. The new BOINC clients return different cpu id information and those old WUs probably can't find anything to match. @David, Pull the chain on those old ones....LOL |
||
|
ID:
2827 | Rating: 0
| rate:
/
|
||
|
Andre Kerstens
Forum moderator Project tester Volunteer tester
Joined: Sep 11 06 Posts: 749 ID: 1 Credit: 15,199 RAC: 0 |
||
|
Hi all,
I believe they found some problems that only showed up when the work queues were drained. IIRC, Andre said that those were very old work units that the system is trying to send out again in order to meet a quorum. My guess would be that getting everything else out of the way is now letting them find the work units that were orphaned by earlier changes in the HR and Quorum rules plus changes in the BOINC client software. The new BOINC clients return different cpu id information and those old WUs probably can't find anything to match. ____________ D@H the greatest project in the world... a while from now! |
||
|
ID:
2828 | Rating: 0
| rate:
/
|
||
|
Hey guys.. just some observations..
|
||
|
ID:
2893 | Rating: 0
| rate:
/
|
||
|
David Ball
Forum moderator Volunteer tester
Joined: Sep 18 06 Posts: 274 ID: 115 Credit: 1,634,401 RAC: 0 |
||
|
Hey guys.. just some observations.. When you were getting the varying times, was that machine running 64-bit windows? BTW, there was an article in Technet about the Vista kernel which said they changed the way they did accounting for CPU time. IIRC, previously whatever process was running when the interrupt occurred to end the time slice got charged for the entire time slice, even if other processes had used the CPU for 99% of that time slice. I wonder if that fix was backported to 64-bit XP? It could also be something in the way the OS handles CPU throttling for powersave or the way it migrates threads between cores. These days, there's so much CPU specific stuff that an OS has to handle that it could be just about anything. I've noticed that, after recent updates from Microsoft, my CPDN machine has gone from taking about 3.9 seconds/timestep to taking 4.1 seconds/timestep. Of course, that's on a celeron 2.3 GHz with 128KB cache, so it's extremely sensitive to cache pollution. Happy Crunching, -- David ____________ The views expressed are my own. Facts are subject to memory error :-) Have you read a good science fiction novel lately? |
||
|
ID:
2895 | Rating: 0
| rate:
/
|
||
|
Yes, and No, this was the same install of 64-bit Windows XP Professional.
|
||
|
ID:
2898 | Rating: 0
| rate:
/
|
||
|
Andre Kerstens
Forum moderator Project tester Volunteer tester
Joined: Sep 11 06 Posts: 749 ID: 1 Credit: 15,199 RAC: 0 |
||
|
Must be something else: the app and workunits have been exactly the same since months now. Also the 32-bit and 64-bit apps really both the same 32-bit app.
Yes, and No, this was the same install of 64-bit Windows XP Professional. ____________ D@H the greatest project in the world... a while from now! |
||
|
ID:
2906 | Rating: 0
| rate:
/
|
||
Message boards : Application Info : charmm_5.4_windows_amd64
Database Error: The MySQL server is running with the --read-only option so it cannot execute this statement
array(3) {
[0]=>
array(7) {
["file"]=>
string(47) "/boinc/projects/docking/html_v2/inc/db_conn.inc"
["line"]=>
int(97)
["function"]=>
string(8) "do_query"
["class"]=>
string(6) "DbConn"
["object"]=>
object(DbConn)#21 (2) {
["db_conn"]=>
resource(72) of type (mysql link persistent)
["db_name"]=>
string(7) "docking"
}
["type"]=>
string(2) "->"
["args"]=>
array(1) {
[0]=>
&string(51) "update DBNAME.thread set views=views+1 where id=186"
}
}
[1]=>
array(7) {
["file"]=>
string(48) "/boinc/projects/docking/html_v2/inc/forum_db.inc"
["line"]=>
int(60)
["function"]=>
string(6) "update"
["class"]=>
string(6) "DbConn"
["object"]=>
object(DbConn)#21 (2) {
["db_conn"]=>
resource(72) of type (mysql link persistent)
["db_name"]=>
string(7) "docking"
}
["type"]=>
string(2) "->"
["args"]=>
array(3) {
[0]=>
object(BoincThread)#3 (16) {
["id"]=>
string(3) "186"
["forum"]=>
string(2) "11"
["owner"]=>
string(1) "1"
["status"]=>
string(1) "0"
["title"]=>
string(24) "charmm_5.4_windows_amd64"
["timestamp"]=>
string(10) "1175488537"
["views"]=>
string(4) "1374"
["replies"]=>
string(2) "15"
["activity"]=>
string(20) "4.8164761171211e-122"
["sufferers"]=>
string(1) "0"
["score"]=>
string(1) "0"
["votes"]=>
string(1) "0"
["create_time"]=>
string(10) "1172529664"
["hidden"]=>
string(1) "0"
["sticky"]=>
string(1) "0"
["locked"]=>
string(1) "0"
}
[1]=>
&string(6) "thread"
[2]=>
&string(13) "views=views+1"
}
}
[2]=>
array(7) {
["file"]=>
string(63) "/boinc/projects/docking/html_v2/user/community/forum/thread.php"
["line"]=>
int(184)
["function"]=>
string(6) "update"
["class"]=>
string(11) "BoincThread"
["object"]=>
object(BoincThread)#3 (16) {
["id"]=>
string(3) "186"
["forum"]=>
string(2) "11"
["owner"]=>
string(1) "1"
["status"]=>
string(1) "0"
["title"]=>
string(24) "charmm_5.4_windows_amd64"
["timestamp"]=>
string(10) "1175488537"
["views"]=>
string(4) "1374"
["replies"]=>
string(2) "15"
["activity"]=>
string(20) "4.8164761171211e-122"
["sufferers"]=>
string(1) "0"
["score"]=>
string(1) "0"
["votes"]=>
string(1) "0"
["create_time"]=>
string(10) "1172529664"
["hidden"]=>
string(1) "0"
["sticky"]=>
string(1) "0"
["locked"]=>
string(1) "0"
}
["type"]=>
string(2) "->"
["args"]=>
array(1) {
[0]=>
&string(13) "views=views+1"
}
}
}
query: update docking.thread set views=views+1 where id=186