Reading the blog post, I was sorry to hear that "performance bottlenecks in the SQL parser” prevented you from making SQL your main interface to Druid. Can you tell us more about the issues? Is there a bug logged against Calcite or Druid?
In the blog post, thanks for qualifying the Druid, Calcite and HBase trademarks with the word “Apache”. Can you please do the same for Hadoop and Kafka? Julian > On Mar 9, 2020, at 4:48 PM, Julian Jaffe <[email protected]> wrote: > > Hey all, > > I recently wrote a proposal <https://github.com/apache/druid/issues/9463> > to add namespacing to Druid segments to allow for more flexible ingestion. > I'm planning on giving it a little more polish in the next few days, but > I'd be very interested in any community feedback. We've been running a > similar system for a year now with great success. > > There have also been a number of people who have reached out after > attending the meetup we hosted or reading our blog post > <https://medium.com/pinterest-engineering/powering-pinterest-ads-analytics-with-apache-druid-51aa6ffb97c1> > and > asked for more details about how we implemented some of our customizations, > and so I'm also hoping that even if there isn't enough community appetite > for this change in Druid, the proposal can serve as a guide for people to > follow if they wish to implement this on their own. > > Julian > > > <https://github.com/apache/druid/issues/9463> --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
