The federation work sounds a bit like what https://drill.apache.org/ does with SQL. Is this the right way to think about it?
Thanks for explaining. Sounds very promising . Btw do you have a link to what PostgreSQL JOB is? I only get career stuff or links about pg_cron, but I feel you are talking about a specific concept within postgres
Ah wait I see JOB means Joint Order Benchmark
Ah wait I see JOB means Join Order Benchmark
I am not familiar with the other solutions, so i will look into them. I was mentioning Drill because indeed at a high level it allows you to query many different data formats with an external SQL query engine. And Datalevin would fulfill a similar use case with a Datalog approach. AFAIK Drill can distribute queries over multiple machines. Is this also something Datalevin will be able to do eventually?
Yeah maybe I'm asking the wrong questions. I'm still trying to understand how/what datalevin federation does exactly. And I just thought of Drill as a reference. I will study the other solutions and reread your blog post to get a more complete mental model