We’re super excited to announce RDB Loader R33 with mutiple bugfixes and improvments, mostly following automigrations added in R32.
Also we’re taking an opportunity to prepare the Shredder for entirely new streaming architecture and removing unpopular features: Postgres (please use Snowplow Postgres Loader instead), Processing Manifest, Hadoop Shred.
Among most important changes:
- EMR 6 and Apache Spark 3
- Big performance improvement in Shredder. Different benchmarks give 1.5-2x speed-up
- Many small bugfixes:
exhausted input error(#203), non-nullable empty string (#171), support of multiple DB schemas (#178)
VARCHAR(65535)for JSON arrays
R33 Upgrade Guide is availablle on our docs website.