A 800 MB catalina logs, with spark analysis and local analysis, the performance gap is a little big
[root @ centos1 TMP] # time cat catalina. Out | grep "error" - c
92

The real 0 m11. 931 s
The user 0 m0. 443 s
Sys 0 m1. 590 s

Also, do the same thing with the spark, with 23 s, because the small amount of calculation, couldn't reflect the ability to spark?

CodePudding user response:

Themselves under the roof,=superior