Splitting databases and opening all of them afterwards does not reduce the memory usage, it might even slightly increase it. As long as there’s enough RAM that’s of course no issue but otherwise it would make things worse.
Smaller databases have of course advantages if all of them are not opened concurrently. In addition, database operations (verifying, optimizing or updating) are faster.
Sorry to throw this a bit off topic, but I also have a question concerning performance.
When I open large MD files on preview mode, I sometimes get the dreaded spinning beach volley-ball for up to 15 secs. Is this also ram related? By “large” I mean files with, let us say, up to 100k words.
Running an iMac 27" 2014 with 32gb ram and a MacBook Pro 15" 2019 with 16gb.
RAM shouldn’t be an issue in this case. Could you please launch Apple’s Activity Monitor application (see Applications > Utilities), choose DEVONthink 3 while it’s frozen in the list of processes, select the menu item View > Sample Process and send the result to cgrunenberg - at - devon-technologies.com? Thanks a lot!
And don’t forget the benefit of syncing less data in smaller databases, which can be especially nice for mobile. e.g., syncing smaller databases for use in the field instead of carrying 90% of what you’ll never need out there!
Sort of related on the topic of splitting: is there a way to see how many words are contained in any group? I can find this easily for each database in ‘properties’ but not in the ‘get info’ per group. Knowing how many words each group contains would be very helpful figuring out where/how to split large collections. I’m blessed with limited storage on my iPad and so am entirely persuaded by your sage advice to run smaller collections for mobile.
@cgrunenberg could comment on whether total words _(i.e. including subgroups) could be an attribute.
It is possible to script something, though the performance would be lessened with more and more documents.
The Word Count column of list views displays the number for groups too.
Amazing, you learn something every day (well, really you learn a lot more than one thing by visiting these forums and its amazing user base, support, and devs!). Thank you :).
True, but only for children of the group, not grandchildren, great granchildren, etc. in subgroups. Perhaps I phrased my statement poorly.
I do encounter the Spinning Beach Ball of Death a bit more often than i would like. I only have one database open at a time (and only one real database, the other is only 16mb) and it’s 40GB with a 489K word count. I’m running a 3.5Ghz 2014 iMac with 24 GB RAM and a Fusion drive (with a 128GB flash rather than the more paltry 24GB in later versions)
If you haven’t yet, one thing to try is “Optimize Database”. My main database is a lot smaller than yours, at 3.4G, with about 10,700 documents and 9.4M words, of which 158,000 are unique. I used to find opening the database quite slow, often with the spinning beachball, but since I optimized the database, opening is basically immediate. So Optimize Database does a great job. This is on a 2017 iMac 27" with 16G memory.
That’s a great tip, thanks!
When I ran into this problem a few months ago, my local system guru spoke magic to my computer and discovered that the cause of my suffering was the sheer amount of disk access I was asking of my machine. He recommended that I upgrade to a system with SSD to eliminate this problem. I bought a quad-core mini; I didn’t buy the biggest SSD, because I didn’t need the space and size actually doesn’t matter here, rather the way the drive operates.
I have had no performance difficulties since.