Details about "thru-Batches"

Message boards : Number crunching : Details about "thru-Batches"
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile Aurel
Avatar

Send message
Joined: 25 Feb 13
Posts: 216
Credit: 9,899,302
RAC: 0
Message 912 - Posted: 29 Sep 2013, 18:58:58 UTC

Hi all,

a little question to all active users and members.
What are you thinking for more informations about the "thru-batches"? (Example: DS10*262 thru DS10*269)

The idea is to list all batches single, from DS10x262 to DS10x269; when we use the example.
The reasen is: I´m an statistic-freak. ;)
ID: 912 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Eric Driver
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 8 Jul 11
Posts: 1318
Credit: 404,017,038
RAC: 289,728
Message 913 - Posted: 29 Sep 2013, 21:23:27 UTC - in response to Message 912.  

Hi all,

a little question to all active users and members.
What are you thinking for more informations about the "thru-batches"? (Example: DS10*262 thru DS10*269)

The idea is to list all batches single, from DS10x262 to DS10x269; when we use the example.
The reasen is: I´m an statistic-freak. ;)



One reason I do that is to reduce the size of the table. Each of those cases has the same discriminant bound, so their runtimes should be comparable. They are also small data sets, so will finish within a couple days. It just didn't seem worth the trouble of cluttering up the table with a bunch of data sets which would finish relatively fast.

Exactly what kind of stats are you looking for? I have tons of stats available after I compile all the results together; for example, number of polynomials tested, number of fields found, and total run times; and it's sorted by case (e.g. 10x262 stats versus 10x263 stats). I could email you whatever you want, just let me know.
ID: 913 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Aurel
Avatar

Send message
Joined: 25 Feb 13
Posts: 216
Credit: 9,899,302
RAC: 0
Message 914 - Posted: 30 Sep 2013, 16:22:45 UTC - in response to Message 913.  

Hi,

you can send me all stats you have. ;)
Ehh, named an good example. "Number of Fields found", I think it´s interesting. (I can make an daily diagramm of fields found/day)

I make an little stats-diagramm over the project-status, so I´ll find it better, when you can send me number of WU´s from every case. ;)

Thanks.

Aurel
ID: 914 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Eric Driver
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 8 Jul 11
Posts: 1318
Credit: 404,017,038
RAC: 289,728
Message 915 - Posted: 1 Oct 2013, 2:00:56 UTC - in response to Message 914.  

Hi,

you can send me all stats you have. ;)
Ehh, named an good example. "Number of Fields found", I think it´s interesting. (I can make an daily diagramm of fields found/day)

I make an little stats-diagramm over the project-status, so I´ll find it better, when you can send me number of WU´s from every case. ;)

Thanks.

Aurel


Unfortunately, the project doesn't keep track of "number of fields found" per day, at least not in an easily accessible format. What I do have is the number of fields found for a given case and how many hours that case took. For example, I might tell you case 10x262 took 2 days and found 200 fields, from which you could infer that 100 fields were found per day. But multiple cases are usually running simultaneously, so I am not sure how useful that information would be.

The database archives contain receive times for all the WUs plus a bunch of other data. So what you seek is contained in there, but you'd have to be crazy to try and sift through all of that data just to get a few statistics. If you were really good at it I guess you could write a shell script to zcat the archives and then grep through that. If you are still interested, I can send you an archive file to play with. Just let me know.
ID: 915 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Aurel
Avatar

Send message
Joined: 25 Feb 13
Posts: 216
Credit: 9,899,302
RAC: 0
Message 916 - Posted: 1 Oct 2013, 6:41:12 UTC - in response to Message 915.  

Hi,

I don´t know, if I can write a shell script so good.

So, I want only know how many WU´s we need to compute an case like DS10x267.
ID: 916 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Richard Haselgrove

Send message
Joined: 28 Oct 11
Posts: 179
Credit: 220,514,402
RAC: 129,251
Message 917 - Posted: 1 Oct 2013, 13:19:05 UTC - in response to Message 916.  

Hi,

I don´t know, if I can write a shell script so good.

So, I want only know how many WU´s we need to compute an case like DS10x267.

I see messages like
01/10/2013 13:44:51 | NumberFields@home | Starting task wu_sf3_DS-10x266_Grp893of902_0 using GetDecics version 102 in slot 6
01/10/2013 13:48:11 | NumberFields@home | Starting task wu_sf3_DS-11x262_Grp549of600_0 using GetDecics version 102 in slot 4

Maybe:

DS-10x266_Grp 893 of 902
DS-11x262_Grp 549 of 600

gives you the information you are seeking?
ID: 917 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Eric Driver
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 8 Jul 11
Posts: 1318
Credit: 404,017,038
RAC: 289,728
Message 918 - Posted: 1 Oct 2013, 23:31:41 UTC - in response to Message 917.  

Hi,

I don´t know, if I can write a shell script so good.

So, I want only know how many WU´s we need to compute an case like DS10x267.

I see messages like
01/10/2013 13:44:51 | NumberFields@home | Starting task wu_sf3_DS-10x266_Grp893of902_0 using GetDecics version 102 in slot 6
01/10/2013 13:48:11 | NumberFields@home | Starting task wu_sf3_DS-11x262_Grp549of600_0 using GetDecics version 102 in slot 4

Maybe:

DS-10x266_Grp 893 of 902
DS-11x262_Grp 549 of 600

gives you the information you are seeking?


That's correct, the total number of WUs in a case is contained in the WU name.
ID: 918 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote

Message boards : Number crunching : Details about "thru-Batches"


Main page · Your account · Message boards


Copyright © 2024 Arizona State University