Charmm 5.04 (Windows)
Message boards : Number crunching : Charmm 5.04 (Windows)
| Author | Message | |
|---|---|---|
|
Just finished my first 5.04 wu. Yes it took about half the time as before on my HT P4, and also claimed half the credit it used to claim. I assume this is what was expected. Nice optimization. Looking forward to when server-assigned credit can happen. |
||
ID:
2024 | Rating: 0
| rate:
/
|
||
|
I'll second that. My Pentium 4/550 with HT (Windows XP) used to take about 10 hours per work unit. I have just finished a work unit with the new app and it only took 4.5 hours. That's actually slightly quicker than it does them under Ubuntu Linux.
|
||
|
ID:
2027 | Rating: 0
| rate:
/
|
||
|
I've just attached my Pentium D (WinXP) to some Docking wu's and will post back when the results come in...
|
||
|
ID:
2033 | Rating: 0
| rate:
/
|
||
|
Windows XP SP2 Athlon 64 X2 4600+
|
||
|
ID:
2036 | Rating: 0
| rate:
/
|
||
|
First two 5.04 wu's are finished now...
|
||
|
ID:
2043 | Rating: 0
| rate:
/
|
||
|
For the records: charmm 5.04 gives invalid results when validated against the previous 5.03 version.
|
||
|
ID:
2044 | Rating: 0
| rate:
/
|
||
Andre Kerstens
Forum moderator Project tester Volunteer tester 
Joined: Sep 11 06 Posts: 749 ID: 1 Credit: 15,199 RAC: 0 |
||
|
Hmmm, I wonder if that is a bug/feature in BOINC; my understanding was that the results in a workunit will always be crunched with the same app version. I guess that was an invalid assumption.
For the records: charmm 5.04 gives invalid results when validated against the previous 5.03 version. ____________ D@H the greatest project in the world... a while from now! |
||
|
ID:
2049 | Rating: 0
| rate:
/
|
||
|
Now I am thinking if the credit per hour between windows and linux is still much higher for windows that for linux. Will look into that and if anyone can provide us with some analysis for their computers it will very appreciated. |
||
|
ID:
2053 | Rating: 0
| rate:
/
|
||
|
Now I am thinking if the credit per hour between windows and linux is still much higher for windows that for linux. Will look into that and if anyone can provide us with some analysis for their computers it will very appreciated. I think it also depends on the manager used at this moment... Two of my hosts: Linux - claim - grant 13,581.94 - 32.07 - 32.27 (BOINC 5.8.1) Athlon Xp Windows - claim - grant 13,658.52 - 30.36 - pending Pentium D (BOINC 5.4.11) As you can see my linux box is claiming a bit higher p/sec at this moment. ;-) ____________ 
|
||
|
ID:
2055 | Rating: 0
| rate:
/
|
||
|
I have also the issue with 5.04 flagged invalid when running against 5.03 in this wu for example. |
||
|
ID:
2076 | Rating: 0
| rate:
/
|
||
|
Wow ! Impressive performance. On my AMD K7/800 from about 19 to 20 hours, now about 8 to 9 hours with 5.04. |
||
|
ID:
2078 | Rating: 0
| rate:
/
|
||
|
Yes, impressive
|
||
|
ID:
2082 | Rating: 0
| rate:
/
|
||
|
> Yes a great improvement
|
||
|
ID:
2083 | Rating: 0
| rate:
/
|
||
|
> Yes a great improvement Memo asked about credit with Windows old app and new app against Linux P4 2.53 @2.75 with Windows v5.03 cr/h average 8.20 in December cr/h average 9.19 (4 WU's) Docking v5.04 4800+ with Windows v5.03 cr/h average 13.83 during December/early January cr/h average 15.08 (3 WU's) with Docking v5.04 Opteron 275 on Linux cr/h average 9.84 Opteron 285 on Linux cr/h average 11.38 >> Andre, Memo and Docking team, > There is definately a problem with the 5.04 app getting lumped together with 5.03 and then not being validated even though successful (it qualifies to give the other two 5.03 WU's their credit even though the 5.04 gets nothing). I suppose this will go away as the 5.03 WU's get done but it will be a problem every time you do an application update. See this result http://docking.utep.edu/result.php?resultid=75000 Thanks and keep up the good work. ____________ 
|
||
|
ID:
2084 | Rating: 0
| rate:
/
|
||
|
suguruhirahara
Forum moderator Volunteer tester 
Joined: Sep 13 06 Posts: 282 ID: 15 Credit: 56,614 RAC: 0 |
||
|
My host has a trouble. One of the tasks which should have been crunched on Charmm 5.03 was distributed on my host, whose Charmm had been already upgraded to 5.04.
|
||
|
ID:
2085 | Rating: 0
| rate:
/
|
||
|
Andre Kerstens
Forum moderator Project tester Volunteer tester
Joined: Sep 11 06 Posts: 749 ID: 1 Credit: 15,199 RAC: 0 |
||
|
Suguru,
My host has a trouble. One of the tasks which should have been crunched on Charmm 5.03 was distributed on my host, whose Charmm had been already upgraded to 5.04. ____________ D@H the greatest project in the world... a while from now! |
||
|
ID:
2086 | Rating: 0
| rate:
/
|
||
|
Have done 2 wu's with the 5.04 app, running Boinc manager 5.8.2 and Win Vista.
|
||
|
ID:
2091 | Rating: 0
| rate:
/
|
||
|
Andre Kerstens
Forum moderator Project tester Volunteer tester
Joined: Sep 11 06 Posts: 749 ID: 1 Credit: 15,199 RAC: 0 |
||
|
Here is David's answer. It seems to be normal behavior of BOINC. We will try to think up a possible solution for this.
Suguru, ____________ D@H the greatest project in the world... a while from now! |
||
|
ID:
2093 | Rating: 0
| rate:
/
|
||
|
Concerning the invalid results with charmm 5.04 validated against 5.03 I would like to thank all fellow crunchers who have finished their results and have not aborted them. Examples:
|
||
|
ID:
2094 | Rating: 0
| rate:
/
|
||
|
Here is David's answer. It seems to be normal behavior of BOINC. We will try to think up a possible solution for this. This explains why SIMAP used to turn off bit-wise validation when they changed app versions. They used to announce that bit-wise validation would be turned off for a certain period, and all crunchers had to complete any "old app" wu's within that period otherwise you wouldn't get credit. I guess the projects that don't have to do this have written their own validator. ____________ 
|
||
|
ID:
2101 | Rating: 0
| rate:
/
|
||
|
Andre Kerstens
Forum moderator Project tester Volunteer tester
Joined: Sep 11 06 Posts: 749 ID: 1 Credit: 15,199 RAC: 0 |
||
|
Will discuss this with the team this week. We will probably have to go the same route (although not for this latest app change, that's too late now). At least we found out about this issue and can do something about it next time :-) Andre ____________ D@H the greatest project in the world... a while from now! |
||
|
ID:
2106 | Rating: 0
| rate:
/
|
||
|
I haven't finished my first 5.04 WU yet, but it does seem significantly faster than 5.03.
|
||
|
ID:
2249 | Rating: 0
| rate:
/
|
||
|
David Ball
Forum moderator Volunteer tester 
Joined: Sep 18 06 Posts: 274 ID: 115 Credit: 1,634,401 RAC: 0 |
||
|
Now to my questions. Should it be checkpointing so frequently? Hello, At one point, Andre said that the application was ignoring the disk interval and checkpointing very frequently while they debug the Charmm application. It looks like that's still the case, even with version 5.04. -- David ____________ The views expressed are my own. Facts are subject to memory error :-) Have you read a good science fiction novel lately? |
||
|
ID:
2253 | Rating: 0
| rate:
/
|
||
|
Andre Kerstens
Forum moderator Project tester Volunteer tester
Joined: Sep 11 06 Posts: 749 ID: 1 Credit: 15,199 RAC: 0 |
||
|
David, that is correct. 5.4 for just for the speed optimization. Probably the next version will do something about the checkpointing interval (Richard is working on this). We still needs lots of debug info though, so the charmm logfile will probably be written to much more often. But that will only be for the alpha test phase.
Hello, ____________ D@H the greatest project in the world... a while from now! |
||
|
ID:
2269 | Rating: 0
| rate:
/
|
||
|
David, that is correct. 5.4 for just for the speed optimization. Probably the next version will do something about the checkpointing interval (Richard is working on this). We still needs lots of debug info though, so the charmm logfile will probably be written to much more often. But that will only be for the alpha test phase. Thanks Andre. Now I remember reading that. That'll teach me to post in the middle of the night.... |
||
|
ID:
2273 | Rating: 0
| rate:
/
|
||
Message boards : Number crunching : Charmm 5.04 (Windows)
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)#30 (2) {
["db_conn"]=>
resource(138) 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=156"
}
}
[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)#30 (2) {
["db_conn"]=>
resource(138) of type (mysql link persistent)
["db_name"]=>
string(7) "docking"
}
["type"]=>
string(2) "->"
["args"]=>
array(3) {
[0]=>
object(BoincThread)#3 (16) {
["id"]=>
string(3) "156"
["forum"]=>
string(1) "2"
["owner"]=>
string(2) "43"
["status"]=>
string(1) "0"
["title"]=>
string(21) "Charmm 5.04 (Windows)"
["timestamp"]=>
string(10) "1169314345"
["views"]=>
string(4) "1840"
["replies"]=>
string(2) "24"
["activity"]=>
string(20) "6.0823791548201e-125"
["sufferers"]=>
string(1) "0"
["score"]=>
string(1) "0"
["votes"]=>
string(1) "0"
["create_time"]=>
string(10) "1168423422"
["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) "156"
["forum"]=>
string(1) "2"
["owner"]=>
string(2) "43"
["status"]=>
string(1) "0"
["title"]=>
string(21) "Charmm 5.04 (Windows)"
["timestamp"]=>
string(10) "1169314345"
["views"]=>
string(4) "1840"
["replies"]=>
string(2) "24"
["activity"]=>
string(20) "6.0823791548201e-125"
["sufferers"]=>
string(1) "0"
["score"]=>
string(1) "0"
["votes"]=>
string(1) "0"
["create_time"]=>
string(10) "1168423422"
["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=156