Phoenix provides a JDBC driver for Java client and hence can be connected to Phoenix by following the steps required to get a JDBC connection. As with JDBC drivers for other DBMS, there are are some Phoenix specific requirements to get a JDBC connection. For a non secure HBase cluster the Phoenix JDBC connection string should be of the form jdbc:phoenix:<ZK-QUORUM>:<ZK-PORT>:<ZK-HBASE-NODE>. The following is the code snippet to get a Phoenix JDBC connection object for a non secure HBase cluster.

If you have been using HBase off-heap bucketcache, you may agree that configuration it is a bit cumbersome to say the least. In version 1.0, the HBase development team simplified the offheap cache configuration process. With the changes, the following are the steps to configure bucketCache for e.g. of size n GB.

A distributed system is a collection of independent computers that appears to its users as a single coherent system. This paper argues that the objects in a distributed object oriented system form a single ontological class where all entities can be described by the specification of the set of interfaces of the objects and the semantics of operation is mistaken. This vision of unified objects for distributed systems is centered around the principles that

This paper presents the design principle regarding placement of functions in computer system design called “end-to-end argument”. The argument of the principle is that any application functions implemented at lower levels of a system may be redundant or of little value when compared with the cost of providing them at lower level. The paper articulates the argument through requirements and examples in distributed systems like reliable data transmission, encryption, duplicate message detection, message sequencing, detecting host crashes, delivery receipts etc.

This paper looks at scale and how it affects distributed systems including highlights of how how scale is addressed in existing systems. A system is said to be scalable if it can handle addition of resources and users without suffering noticeable loss in performance or increase in administrative complexity. Scale also affects the way users perceive the systems. For e.g. as the number of objects accessible grows it becomes increasingly difficult to locate the objects of interest.