I didn't take a deep look at this, but the general idea of setting up a worker queue as a filesystem structure & synchronizing via sync-thing is pretty cool.
Only question is, would this support multiple-workers? Would be neat to be able to distribute work a to a cluster this way.
2 comments
[ 3.4 ms ] story [ 13.3 ms ] threadOnly question is, would this support multiple-workers? Would be neat to be able to distribute work a to a cluster this way.
https://www.complete.org/parallel-processing-of-filespooler-... discusses parallel processing in Filespooler. This isn't a design goal and may not be the best fit, but when everything's just a file, some tricks are possible here.
Another interesting use case is one-to-many and many-to-many; https://www.complete.org/filespooler/ describes those.