Oracle customers have long been familiar with clustering. As the market leader in Highly Available databases the pedigree Oracle has in this area is undeniable, with Real Application Clusters (RAC) being one of the most prevalent Database Clustering solutions ever. However, database clusters are only part of the story.

Why Cluster?

Clustering serves multiple purposes;

•Achieve High Availability (HA) by eliminating any hardware as a single point of failure,
•Allows scalability beyond the physical limits of a single server.
•Provides capabilities for consolidation to reduce costs and improve manageability
Clustering at any Tiers

In today’s demanding environment High Availability and Performance are no longer seen as mutually exclusive capabilities. Any company, small or large, demands both without compromise, and therefore clustering is becoming a key part of system architectures.

Clustering has also evolved to go beyond the database tier, with clustering at the Application Server as well as in memory data clustering. e-DBA can help organizations decide on the right cluster methodology to support their business including:

• High Availability and Disaster Recovery Planning
• In Memory Data Grid Architecture
• Loosely coupled clustering
• Data Grid/Compute Clustering for analytics or high performance computing.
• Reference Architectures

Getting it Right

Whilst it is un-deniable that with the advent of commodity computing clusters have got easier, both to install, configure and support, it is still essential to architect the solution properly and ensure it is built to best practice, this is where e-DBA comes in.

Building on our extensive experience we can help architect the solution to support your requirements, both at hardware and software levels, as well as on-going support. With experience in Financial Services, eGaming and Retail we can ensure that we bring our skills to bear. Not only helping you architect but benchmark, test, tune and maintain.

Whether you’re looking at Highly Availabl databases, scaling out an existing application, building an in memory compute cluster, or map-reduce technologies we can advise on the most appopriate mechanisms.