spark-bench
spark-bench copied to clipboard
The querytime is incorrect since it is lazy evaluated
Now the source code of spark-bench uses single scala time function to evaluate the execution time.
The time of querytime is incorrect if we just call the sql function which will return right now.


Kind of similar with https://github.com/CODAIT/spark-bench/issues/161 .
Querytime is the most important part here. Does this bug also affect other Sparkbench workloads?