GitHub user shadowmmu added a comment to the discussion: [Performance] Velox Bloom Filter Inefficiency vs. Photon at 1TB Scale
And on our DBX we have disabled all sorts of caching (with whatever config options we could have found), and enabled CBO for spark as well. The major difference we have found are : highly effective bloom filter with extended limit and efficiency, Broadcast hash join, different and more optimized query plans, Aggregation is much faster then Velox etc. GitHub link: https://github.com/apache/incubator-gluten/discussions/11554#discussioncomment-15686096 ---- This is an automatically sent email for [email protected]. To unsubscribe, please send an email to: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
