Some key features and improvements in this release:
1. Optimized scan performance by avoiding multiple data copies and
avoided double filtering for spark fileformat and presto by lazy loading
and page pruning.
2. Added more CLI enhancements by adding more options.
3. Supported fallback mechanism when offheap memory is not enough then
switch to onheap
4. Enable Local dictionary by default
5. Supported C++ interface for writing carbon data in CSDK
6. Supported concurrent reading through SDK reader to improve read
7. Supported VectorizedReader for SDK Reader to improve read performance.
8. Supported customize column compressor
9. Make inverted index false by default
10. Support prefetch for compaction to improve compaction performance.
11. Supported a separate audit log.
12. Many Bug fixes and stabilized Carbondata.