As we automate more and more business critical applications; keeping those applications available becomes more important.
If you’re using eCommerce, how much does it cost your business when your eCommerce application is unavailable? If you’re in a hospital or other medical setting, what are the consequences when Electronic Medical Records are inaccessible? If you manage a utility, what happens when a system that manages the flow of electricity, water or E911 calls is unavailable?
We need to plan for the unavailability of all of our systems, and, in some cases, even an outage of only a few seconds could have life-threatening consequences. On the other hand, even at period-end, if an internal accounting system is unavailable for an hour, the consequences are much smaller and more manageable than if Electronic Medical Records, or a system managing a nuclear power plant are unavailable.
How much high-availability is enough? How do we minimize the cost of providing high-availability? How do we make sure that our high-availability systems work when we need them?
The current minimum standard for high-availability for an Oracle database is Oracle Real Application Clusters. However, even with RAC, there’s a “single point of failure” at the database level.
A step beyond this would be Oracle’s Autonomous Database, which comes with a 99.995% or less than 2.5 minutes of downtime per month availability “guarantee”. In some cases, even that 2.5 minutes per month could be problematic; but, in others, is it worth the cost? What about your “legacy” applications? Is it worth the cost and effort required to move those to “The Cloud”?
Using our award-winning Database Replication Product, SharePlex; you can achieve high levels of availability at a reasonable cost. In my next blogs and in this WebEx, I’ll show how SharePlex can be used to compliment RAC, or achieve “good enough” High Availability, even without RAC.