Github user MaxGekk commented on the issue:
https://github.com/apache/spark/pull/21909
> does this benchmark result vary if we select col2 or col10?
@felixcheung Not so much. Here is the benchmark for CSV.
```
JJava HotSpot(TM) 64-Bit Server VM 1.8.0_172-b11 on Mac OS X 10.13.6
Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz
Count a dataset with 10 columns: Best/Avg Time(ms) Rate(M/s)
Per Row(ns) Relative
------------------------------------------------------------------------------------------------
col0 + count() 9097 / 9167 1.1
909.7 1.0X
col2 + count() 9294 / 9302 1.1
929.4 1.0X
col5 + count() 9346 / 9394 1.1
934.6 1.0X
col7 + count() 9227 / 9231 1.1
922.7 1.0X
col9 + count() 9141 / 9233 1.1
914.1 1.0X
```
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
