Adds batch processing support #590
Conversation
|
Hey @mhenrixon, thanks for this! I haven't looked at it yet, but there's a previous PR for batch support: #142 and a discussion there. It's work in progress as well, but maybe the discussion there is relevant for this, too. |
This builds on some work that I did for Sidekiq actually. I had to put a database in front of sidekiq to make it work like I wanted and to ensure that no duplicates are processed while some stuff needs to happen at the end.
Hey @rosa, so I had a look at the other PR! Admittedly, I am biased, but I feel that my suggestion is more in line with the code I already saw in solid_queue. I don't care either way as long as the functionality ends up in a release in the not so distant future, though. Would @jpcamara be open to share notes and get the feature across the finish line? I am more than happy to support you in finishing up the work in your branch. |
This builds on some work that I did for Sidekiq.
I had to put a database in front of Sidekiq to make it work as I wanted and to ensure that no duplicates are processed, while some tasks need to occur at the end.
There are several factors to consider, including the tracking of pending jobs. This might be better as a simple query, depending on how fast the jobs are processed; it could cause side effects with concurrent increments.
This is intended to initiate a discussion. There are other ways of handling this, but none that I like. This is the only approach that I can think of that doesn't have too many negatives.
The only negative is the database changes.