Google Cloud BigTable does not support coprocessors:
Co-processors are not supported. You cannot create classes that implement the org.apache.hadoop.hbase.coprocessor interface.
https://cloud.google.com/bigtable/docs/hbase-differences
I understand that coprocessors require the deployment of client code (cans) on each tablet (RS) node. However, Endpoint coprocessors are vital for HBase applications to ensure data locality in some scenarios. HBase extensions like Apache Phoenix rely on Observer coprocessors to support secondary indexes, so the lack of coprocessor support looks like the main area of incompatibility with me.
Is coprocessor support possible in the future and what are some workarounds for executing custom Java stored procedures on BigTable tables?
UPDATE 1: Apache Phoenix Compressor List:
- GroupedAggregateRegionObserver
- indexer
- MetaDataEndpointImpl
- MetaDataRegionObserver
- ScanRegionObserver
- SequenceRegionObserver
- ServerCachingEndpointImpl
- UngroupedAggregateRegionObserver
source share