I was wondering if anyone has an optimization tricks (if it’s even possible) to optimize database syncs between computers and mobile?
In my case I’ve setup multiple databases (for different directories/use cases) and some are quite large indexing 1000s of files. I’ve noticed that sync (even direct on a fast network) can take 5-10 mins on the larger databases which I find odd since the databases are small in size, but I imagine large index files. I assume this means DT is syncing each record vs. the entire table as a blob, or checking each row vs. comparing differences and just doing that. This problem appears to be worse between a desktop and mobile.
I’ve got sync setup as direct, iCloud and to a local store (on a NAS), so plenty of methods in play.
Is there anything I can do to optimize sync processing time?
The initial sync might require some time but afterwards only changes are synchronized. How large are the files actually and how fast is the used network?
There are very few files being synced. The vast majority is nothing but index items in various folders (combo of personal cloud & commercial cloud tools). I keep very little stored on any single machine, since most of my life is on the go. So everything has to be accessible remotely.
The few files that are in my databases (only because I have yet to move them into their respective folders) are usually under 3 megs. Generally PDFs, images, and office docs. And overall the total in any single database is under 100 megs. Again usually I move them into their respective folders so they live in my personal cloud. So what I’m usually syncing are indexes (which the biggest indexed folder has thousands of docs in it) or web-archives and bookmarks.
Network speeds vary from coffee shop/airplane/random wifi (minimal time), ultra high speed office (majority of time) and high speed home (300M at home). When at home my personal cloud replicates down to local NAS storage which runs over a 1G home backbone and AC wifi.
I’m wondering if the issue is the sync of the metadata 1000s of files. If sync is incremental, than you’re probably right, this is just because first sync has yet to complete and then in the future it’ll be faster.
Though does DevonThink have any brains over which source to sync from if multiple are setup? Meaning does it just grab the first it finds, or does it check through the list and pick the fastest? Example: I’m sitting at home so local, direct, and iCloud should all be available, but does it grab iCloud instead of direct just because?