![]() |
#671
|
|||
|
|||
![]()
Sorry, the site is back up now.
Its in our plans to have a backup server (different IP) with automatic rollover, we hope to have that going within a couple of weeks. -Ben |
#672
|
|||
|
|||
![]()
[ QUOTE ]
[ QUOTE ] pt aggregator is importing at about 1000 hands/hour, this is going to take weeks at this rate, I have no other programs running at the same time, any way to speed this up? I read someone mention vacuming your database or something earlier in this thread but I have no idea what that means. [/ QUOTE ] Wow, it really shouldn't be that slow. Try opening up Task manager, from View Select columns add "IO Write Bytes" and "IO Read Bytes", and sort by those columns to see if some other process is using up your disk resources. Another thing to try is to reboot, just in case some background processes are taking up some resources. Let me know if that helps at all. -Ben [/ QUOTE ] Well, I restarted my comp, ran adaware, and now it's going at about 8,000 hands/hour but this is still going to take forever like this. I opened tast manager and postgres is using 95% of my cpu, but I couldn't figure out how to check the "io write bytes" I really dont feal like not playing/mining for a week so if there's anything I can still do to speed this up i'd like to hear it. |
#673
|
|||
|
|||
![]()
Backup server will be good
I just signed up for 3 months Great product. |
#674
|
|||
|
|||
![]()
[ QUOTE ]
Well, I restarted my comp, ran adaware, and now it's going at about 8,000 hands/hour but this is still going to take forever like this. I opened tast manager and postgres is using 95% of my cpu, but I couldn't figure out how to check the "io write bytes" I really dont feal like not playing/mining for a week so if there's anything I can still do to speed this up i'd like to hear it. [/ QUOTE ] 8,000/hour is still too slow. Do you know how much RAM your system has? I have 768 MB, CPU Speed is 2.8 GHz, and I can do about 100,000 hands/hour. Shutdown PTAgg, go into PokerTracker, Utilities Menu, Postgres database functions, select "Vacuum Full/Analyze" and press Compress. Do these for each of your PT databases. When that's done, also try running Window's Disk Defrag. Let me know if that helps. -Ben |
#675
|
|||
|
|||
![]()
[ QUOTE ]
Well, I restarted my comp, ran adaware, and now it's going at about 8,000 hands/hour [/ QUOTE ] me too importing into the ptagg database has gotten really slow. Like 10K-15K hands per hour after a fresh reboot and running no other programs. My pc has 4GB of RAM. I have vacummed the ptagg database almost everyday. I aggregated one mining database of 2.2Million hands. Now I'm aggregating my second db-my main playing databse for the last year about 500,000 hands. I wanted to aggregate one last mining database of ~2.50 Million hands. It seems to be slowing way down as aggregate more data despite the precautions I'm taking. Its just time-consuming that's all. averaging about 16K per hour now after another reboot. The memory usage starts at ~100K and climbs up to ~400K afer 6-7 hours-definite memory leak. |
#676
|
|||
|
|||
![]()
[ QUOTE ]
[ QUOTE ] Well, I restarted my comp, ran adaware, and now it's going at about 8,000 hands/hour but this is still going to take forever like this. I opened tast manager and postgres is using 95% of my cpu, but I couldn't figure out how to check the "io write bytes" I really dont feal like not playing/mining for a week so if there's anything I can still do to speed this up i'd like to hear it. [/ QUOTE ] 8,000/hour is still too slow. Do you know how much RAM your system has? I have 768 MB, CPU Speed is 2.8 GHz, and I can do about 100,000 hands/hour. Shutdown PTAgg, go into PokerTracker, Utilities Menu, Postgres database functions, select "Vacuum Full/Analyze" and press Compress. Do these for each of your PT databases. When that's done, also try running Window's Disk Defrag. Let me know if that helps. -Ben [/ QUOTE ] 512 in ram, cpu is 1.8ghz I believe, I'll do the vacume and see if that helps, also both databases i'm trying to import are access, could this be what's slowing it down? |
#677
|
|||
|
|||
![]()
[ QUOTE ]
512 in ram, cpu is 1.8ghz I believe, I'll do the vacume and see if that helps, also both databases i'm trying to import are access, could this be what's slowing it down? [/ QUOTE ] It could be that its just slower importing Access databases. For my testing I only tried importing small Access databases, and large Postgres databases. Maybe you could try converting your Access DB to Postgres using the PT menus? I'm not sure how long that would take. -Ben |
#678
|
|||
|
|||
![]()
[ QUOTE ]
[ QUOTE ] Well, I restarted my comp, ran adaware, and now it's going at about 8,000 hands/hour [/ QUOTE ] me too importing into the ptagg database has gotten really slow. Like 10K-15K hands per hour after a fresh reboot and running no other programs. My pc has 4GB of RAM. I have vacummed the ptagg database almost everyday. I aggregated one mining database of 2.2Million hands. Now I'm aggregating my second db-my main playing databse for the last year about 500,000 hands. I wanted to aggregate one last mining database of ~2.50 Million hands. It seems to be slowing way down as aggregate more data despite the precautions I'm taking. Its just time-consuming that's all. averaging about 16K per hour now after another reboot. The memory usage starts at ~100K and climbs up to ~400K afer 6-7 hours-definite memory leak. [/ QUOTE ] To be honest I haven't tested myself on very large db's, just my db with about 150k hands. It might be that it just takes a long time at that size. Like you noticed, there is a memory leak, so restarting it every 4 hours will help too. I'm going to release a new build that lets you change some of the PTAgg settings, this might help speed it up. One setting is the "Batch Size", its currently 100 Hands per batch. It might be that changing this to 500, 1000, etc will speed things up. It will probably take longer to do the one batch, but if its more efficient then overall it will be faster. -Ben |
#679
|
|||
|
|||
![]()
I'd like it if you could put the different limits I'm interested in on different tabs. Like firefox tabs. So I could pick out my 600 tables, then my 400 tables, etc...
It's very jumbled together right now. (Still good though) Krishan |
#680
|
|||
|
|||
![]()
Ben, when I was using it there was absolutely no speedup with the aggregated table. My big db is over a million hands, also have a second much smaller db.
However the main slowdown was when PT was importing to postgresql at the same time. That absolutely kills it, and guess not much you can do about it. For those experiencing slowdowns, I would suggest suspending mining or at least importing observed tables while you play and that should help a lot. |
![]() |
|
|