[GitHub] spark issue #18174: [SPARK-20950][CORE]add a new config to diskWriteBufferSi...

[manku-timma]

Github user manku-timma commented on the issue:

    https://github.com/apache/spark/pull/18174
  
    Just to understand what is happening.
    
    1. Shuffle records are written to a serialisation buffer (1M) after 
serialisation
    2. The serialised buffer is written to in-memory-sorter’s buffer
    3. once in-memory sorter’s buffer is full, the data is copied to 
sorter’s disk buffer (1M)
    4. the sorter’s disk buffer is written out to a buffered output stream 
(buffer = 32k)
    
    I am guessing reducing the sorter’s disk buffer (in step 3) is helping 
because it triggers fewer writes at the step 4.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org