opteryx
opteryx copied to clipboard
mabel-dev
🦖 A SQL-on-everything Query Engine you can execute over multiple databases and file formats. Query your data, where it lives.
In testing 100m literals takes about 8 seconds to build, this is wasted when we only use the first item. We either need - if in all the evaluations for...
### Thank you for helping maintain the quality and performance of Opteryx. _We appreciate your efforts — your contribution ensures that Opteryx remains reliable and efficient for everyone._ --- **Is...
Burn the rows in the reader when the offset is pushed - although offsets will usually be used with ORDER BY clauses so this may not happen often IRL
### Thank you for helping maintain the quality and performance of Opteryx. _We appreciate your efforts — your contribution ensures that Opteryx remains reliable and efficient for everyone._ --- **Is...
Absolutely, caching compiled expressions is not only feasible but would be a game-changer for performance, especially for repetitive queries. Here's how it could work: --- Caching Strategy: 1. Expression Hashing:...
Rather than relying on SUM of an expression, count_where could be optimized for this scenario. This should be invoked with COUNT (*) FILTER (n) syntax This should be faster as...
we can split large workloads using shuffle, need to be able to spill. This would change sorting, we would do a merge join
a null column can probably just be dropped and it will be added again by the concat tables or the exit operator.
~~~sql SELECT salary, COUNT(*) FROM employees GROUP BY RANGE(salary, 10000); ~~~ Groups by bins of 10000 ~~~sql SELECT event_time, COUNT(*) FROM events GROUP BY INTERVAL(event_time, '1 hour'); ~~~ Groups by...