Lemmy Server Performance
lemmy_server uses the Diesel ORM that automatically generates SQL statements. There are serious performance problems in June and July 2023 preventing Lemmy from scaling. Topics include caching, PostgreSQL extensions for troubleshooting, Client/Server Code/SQL Data/server operator apps/sever operator API (performance and storage monitoring), etc.
a large part of the queue going unbounded is due to the retry queue and missing checks if the receiving servers are actually available. quick fix is disabling the retry queue which is currently making it not go unbounded on lemmy.world
storing the queue persistently is somewhat of a separate issue since that doesn't much affect whether or not it can be processed in time.
also a ton of the memory use was (and is) due to inefficient sql queries.
I guess that works as an emergency measure. Persistent storage doesn't affect whether the updates are processed in time, but it would act as a sort of swap to keep the memory usage manageable.
For scalability, perhaps, you could run dijkstra and route the updates using the shortest path to each federated node, in a multicast sort of way? That would make the updates scale in a O(log(N)) way, provided that activity isn't too centralised. It would also be great to run periodic "deep scrubs" between instances to sync up each other's activities and provide actual eventual consistency. I guess that's kind of a liberal interpretation of ActivityPub, but I think that's the only way to ensure real scalability.
Using Couchbase as your eventual consistency database is perfect for this scenario. It's designed for this type of thing. Even if systems are offline for a few days they will queue up and replicate when they come back online. Cruise ships use it for this very reason.
Can you provide a link? Only thing I see on Couchbase is for NoSQL databases.
Actually XDCR is not available in the community edition so that option is gone. Sorry about that.
It is a NoSQL database but the SQL syntax is ANSI SQL compliant. If you moved the queues to Couchbase and let it handle the replication and consistency, you wouldn't have to code for it.
Thanks for checking.
We don't need full database replication. We are just replicating activities that other servers are subscribed to. So the Comments table, only some of the rows will be "replicated". Not sure if/how CouchBase handles this.
Buckets are like databases in traditional relational databases. Here's the doc on how to filter which documents get replicated.
https://docs.couchdb.org/en/stable/replication/intro.html#controlling-which-documents-to-replicate
Normally with XDCR you can specify which documents to replicate out of a bucket. It doesn't have to be the entire bucket. So if you had certain types (comments, upvotes, etc) then only those would sync when the target comes online.
I did check into Apache CouchDB, the open source upstream, and replication is there. We use Enterprise Couchbase at work and it's a dream but there are some tools that I use that only use Apache CouchDB (Inkdrop for example). It's worth looking into.
Sorry. I didn't realize XDCR was not available in the community edition.