I've developed a HBase application which now I need to port to cloud-big-table. Can I use the easy way out of using BigTable Hbase API or should I redo the code to use BigTable API, given reliable high throughput is a prime requirement ? Or what are the more tangible differences between the two APIs in terms of reliability and performance which can help me decide?
Thanks.
Both APIs will give you the performance and reliability you're looking for. Since you're an existing HBase user, I'd recommend using the HBase client so you can get started on Bigtable right away with a lift-and-shift.
The Bigtable Java API was created to provide a consistent experience with other Google Cloud products such as Spanner, Firestore, etc, so if you were using other GCP products, then it's easy to learn the Bigtable Java API since it follows a familiar structure.
To summarize, each API is designed in a way to make it easy for certain users to get onto Bigtable with an interface they are comfortable with and shouldn't have an effect on the performance, throughput, or reliability.