Building a Near High Availability Oracle Database Farm Using iSCSI and Windows Servers

Steve Remsing, Senior Research Specialist, The Dow Chemical Company

Many applications are extremely important to business operations but are not mission critical.  For these applications full high availability solutions are excessive due to the expense and complexity.  This presentation describes how we used commodity hardware and software to build a near high availability database server farm.  For our environment, we see near high availability as: less than ten minutes of downtime for a planned event and full recovery from an unplanned catastrophic failure in less than an hour.  Based on our testing we are well within our recovery targets.  Our solution has increased our application availability using our existing hardware and software with minimal additional expense or support costs.  

As a result of participating in this session, attendees will be able to:  

• Determine when it is appropriate to implement a near high availability solution versus a full high availability solution
  
• Compare and contrast different products or solutions for improving availability for Oracle databases
  
• Implement a solution to meet your business needs.  The method described will work with iSCSI or Fibre Channel SANs on Windows or UNIX/Linux servers