January 2019
Beginner to intermediate
154 pages
4h 31m
English
Your choice of file format can also impact the performance of your Spark applications. You should consider three parameters while choosing the file format for your application:
Binary file formats improve both storages and network transfer. The binary format has a higher rate of compression compared to a text-based file format such as CSV and JSON. You should always choose a splittable file format. This allows different tasks to read a different part of the data in parallel.
The choice between columnar file format and row based file format depends upon the access pattern of the data. If your access pattern suggests frequent read and computes only ...
Read now
Unlock full access