Tuning the Hive Engine for Big Data Management®

Tuning the Hive Engine for Big Data Management®

Case Study: HParser vs. Integrated Data Transformation

Case Study: HParser vs. Integrated Data Transformation

The mapping reads a file using complex file reader, parses and splits the file into three groups, and writes to different targets based on filter conditions.
Nine map tasks were involved in reading approximately 500MB of file with 64 MB block size.
The following image shows the mapping:

Result

HParser was observed to perform better. Big Data Management was approximately 2.57X slower in this specific case.

0 COMMENTS

We’d like to hear from you!