lhotari commented on PR #24346:
URL: https://github.com/apache/pulsar/pull/24346#issuecomment-3077784416

   > Basically, the motivation is wrong, how can you say it's a clear 
argumentation? The root cause is that **there is no test or code example**, so 
the explanation might not match the actual behavior.
   
   @BewareMyPower Sure, that's true. There's room for improvement. Thanks for 
looking into this.
   
   Another way to describe the motivation is to avoid extra latency due to 
sequential reads.
   One different way to look at this PR is that it will fix a performance 
regression that PendingReadsManager could introduce. due to the sequential 
reads when concurrent reads overlap. Before PendingReadsManager and 
InflightReadsLimiter were introduced, the reads would go directly to BookKeeper 
without additional backpressure from what dispatcher does. That's why I'd 
consider the risk of this change causing anything but positive impacts by 
reducing latency.
   
   
   


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to