How To Achieve Facebook “Like” High Availability in the Cloud

When IT organizations start being compared to Facebook, in terms of application availability, it is clear that business application owners are becoming increasingly less tolerant of ANY application downtime. In fact during a recent Storage Swiss webinar, “HA and DR Must Haves For Cloud Based MS SQL”, lead analyst George Crump related a real-life story about an application owner who charged into his CIO’s office to complain about a recent application outage. The CIO gently reminded this person that this had been a planned outage, however the application owner retorted in frustration, “Well, Facebook never goes down!” Thus is the conundrum that many IT organizations now face; the need to provide “always-on” application services without a corresponding increase in funds to provide such services.

Traditional ways of implementing application high availability typically consist of deploying some kind of shared storage solution, like a SAN, across clustered servers that run a mirror image of the application. Production data is then synchronously or asynchronously replicated to an offsite facility to another standby host, to protect against an outage in the primary data center site. The challenge with this approach is that it is costly to implement. SAN infrastructure can easily eclipse six figures and dedicated offsite data center space and infrastructure can double the total cost of ownership of the environment. Secondly, these types of solutions are complex to deploy and manage. Lastly, shared storage is a single point of failure. If an outage occurs on the SAN and their is no offsite data center to fail-over to, then the business will incur downtime until the problem on the SAN is isolated and repaired.

To achieve Facebook “like” levels of HA without committing to inordinate capital outlays, businesses need HA clustering solutions which can leverage either SAN or “SANless” commodity, server-side disk resources for local application fail-over. In fact, by utilizing separate internal disk resources across two hosts in an HA cluster, businesses would benefit from lower cost storage while also eliminating the single point of failure of a shared storage environment.

But as importantly, businesses also need a way of leveraging the cloud to provide a cost-effective means of instantaneously recovering from a primary site application outage. But here again, many traditional HA clustering solutions which rely on access to shared storage fall short because cloud storage cannot be shared across disparate hosts. Workarounds to this issue typically require additional software tools which only increase costs and complexity. To learn how to implement a simple and cost efficient way to achieve application HA for local fail-over and DR in the cloud, go to the Storage Swiss on-demand webinar, “HA and DR Must Haves For Cloud Based MS SQL”.

By accessing this on-demand webinar, you will gain access to more than 100 Storage Swiss webinars. And once you register with our BrightTalk channel, you will never have to re-register for a future BrightTalk event.

Watch On Demand

Watch On Demand

Click Here To Sign Up For Our Newsletter

As a 22 year IT veteran, Colm has worked in a variety of capacities ranging from technical support of critical OLTP environments to consultative sales and marketing for system integrators and manufacturers. His focus in the enterprise storage, backup and disaster recovery solutions space extends from mainframe and distributed computing environments across a wide range of industries.

Tagged with: , , , , , , ,
Posted in Blog

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 22,209 other followers

Blog Stats
  • 1,529,149 views
%d bloggers like this: