[GitHub] [arrow] Dandandan edited a comment on pull request #9588: ARROW-11799: [Rust] fix len of string and binary arrays created from unbound iterator

[GitBox]

Dandandan edited a comment on pull request #9588:
URL: https://github.com/apache/arrow/pull/9588#issuecomment-790962611


   @yordan-pavlov 
   Would be very interested in a faster parquet reader.
   Was doing some benchmarking with some queries in DataFusion with Parquet, 
looks like there is a lot to win in Parquet when looking at profiling results. 
For example, query 5 showing about 50% of instruction fetches are for reading 
parquet:
   
   
![query5](https://user-images.githubusercontent.com/163737/110033545-d54af780-7d39-11eb-945c-3614e3d85c8c.png)
   
   And reading all the parquet files of the benchmark to memory (and running 
some queries) (left inclusive % instruction fetch, right self %).:
   
   
![unknown](https://user-images.githubusercontent.com/163737/110033770-1a6f2980-7d3a-11eb-9c54-1968fe35b88a.png)
   
   


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org