Ram drive/symlink for guide image folder & /or WMC DB

[prsa01]

Topic is probably WELL beyond the scope of this forum but figured I'd put it out there and see if it's even worth messing with and/or other ideas to try first.

System - 6th gen i7 desktop, 16gb mem 256 gb ssd; 12 tb external recording drive
Guide will sometimes slow down to unusable; most recently rebuilding the db via epg123 along with cleaning out the guide image cache folder resolved the issue.
To the best of my knowledge garbage collection should be running when/as intended. Any way I can confirm this?

Interestingly, the previous win7 system I was on for years that this system replaced with less horsepower (mobile i7 3rd gen, 12gb mem, similar ssd) did not seem to have this occasional issue

================

DDR4 RAM is cheap enough via CL/FBM that it has made me consider whether this is potentially worth messing with or if I just need to rebuild my DB anytime the guide starts slowing down. Guide

Guide Details - OTA only, ballpark 25 channels, carrying 10 days guide data
guide image cache folder (ssd) holds about a gig of images
wmc DB size (will add) <my assumption is that guide is in db but images are fetched realtime, is that correct?>

With a ramdrive, I'd need to repopulate it at boot, either use symlinks or fix the guide image paths and probably use a symlino for the wmc DB/ehome.

Anyway, if this isn't really seem worth discussion, the thread can just be removed.

[Space]

I also have slowdowns after a while and have to rebuild the WMC DB. It makes the DB much smaller and faster. I'm not sure that the garbage cleanup is doing it's job (at least on my system). It is running, but I still think it has old data in there (as can be seen by looking at the guide data of the past using Remote Potato).

I watch as the DB keeps getting bigger and bigger, over 1 GB in size and eventually I have to rebuild it to make WMC quick again.

Note that clearing out the image cache for EPG123 should not be necessary, just rebuilding the WMC DB should be sufficient. The WMC DB just has URL links to the images in the cache. I think the images in the cache expire if not used for a while, so the cache should be self maintaining, but I'll let Gary give the details of that.

I would assume a RAM drive would make it a lot faster for longer, but that still doesn't fix the root issue of the DB getting endlessly bigger and eventually I would assume you would have the same issue.

[garyan2]

It does sound like a database problem. You can verify if garbage collection is running by checking the epg123 log file. There will be entries for "Entering PerformGarbageCleanup()" and "Exiting PerformGarbageCleanup()". That should be happening every 5 days to maintain the database. If/When it slows down again, take a look at the database size... that should give you an indication if that is the problem.

Space is right. Clearing the image cache does nothing for you other than force you to download some images again.

[prsa01]

Thanks guys! I should have specified that when I said "cleaning out image folder" I only cleared outvery old images.

I'll check the logs next time the guide slows down and report back and address the issue based on those results.