mcvsubbu commented on pull request #7267: URL: https://github.com/apache/pinot/pull/7267#issuecomment-895624138
> What's the scenario where this feature will be useful? > > Any drawbacks? Looks like this will make entire server not accessible during catch-up instead of just one segment. Is my understanding correct? This is useful in scenarios where there is a very high ingestion rate and low latency requirements for queries (as well as high query throughput). We need to do some catching up during server restarts before we declare the server ready to serve queries. Right now, there is a (configurable) hard time limit that the server waits for before declaring itself ready to serve queries. We would like that the server automatically detect when it is ready to serve queries. The default behavior (of allowing queries right away) will still remain. Once we analyze a few use cases, we will add appropriate configs, so that the feature can be turned on for tables such as those declared above. We were thinking of per server config, but we could make it a table config, so that if it is not configured, the segments declare as being ready right away. The Config part will be a separate PR, and up for discussion. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: commits-unsubscr...@pinot.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@pinot.apache.org For additional commands, e-mail: commits-h...@pinot.apache.org
