Building production ready Phoenix Query Server

With the advent of Machine Learning and AI technologies, NoSQL databases aka non-relational data stores are the need of the hour. Given SQL has been popularly used to query underlying databases, Phoenix — A SQL layer on top of HBase brings the best of both worlds. And the cherry on the top: the Phoenix Query Server(PQS) provides HTTP endpoints for interaction with Phoenix and HBase. This will enable access from environments other than the JVM. Phoenix 4.4 introduces a stand-alone server that exposes Phoenix to “thin” clients. It is based on the Avatica (https://calcite.apache.org/avatica) component of Apache Calcite (https://calcite.apache.org/). The query server is comprised of a Java server that manages Phoenix Connections on the clients’ behalf. The client implementation is currently a JDBC driver with minimal dependencies. Two transport mechanisms are current supported: JSON and Protocol Buffers (available as of Phoenix 4.7 as the default).

In this session, we will talk about available customizations in PQS, production use case of PQS @Salesforce, and orchestrating PQS in a realtime cluster. We will discuss load balancer architecture to leverage PQS and also shed light upon opensource artifact management and release cycle for PQS. Attendees will learn to run PQS and orchestrator locally via a short demo.

Suggested main topic
Understanding PQS in terms of usability
What does PQS provide with respect to metrics and security
Customizing metrics and security mechanisms
Deploying PQS in production
Orchestrating PQS

Audience information:
Who the presentation is for?
Customers planning to use PQS in production
Enabling JVM and non-JVM based applications for PQS
What will they be able to take away?
How to customize PQS for specific metrics and security
How to orchestrate PQS in production
What prerequisite knowledge do they need?
TCP/HTTP/authorization and authentication protocols
Serialization methods - Protobuf and JSON
Basics of Apache Phoenix / HBase