Check out our Product Guide for more info on our highly available software.

A few weeks ago I announced a new blog series on Advanced Rendering and which technical features enterprise organizations need to succeed. This week I’ll talk about the second of the 10 important items: high availability.

Have you been on a plane recently? How important is it that the engine not stop working?

Believe it or not, there are a number of IT examples where the services and applications deployed must never stop operating, much like the jet engine. Having the resource constantly operating is referred to as it having high availability.

Dealing with downtime in the enterprise

What is the cost to your organization when your critical business application is down or unusable? We’ve heard lots of different scenarios from our customers, but many estimate the cost of not being able to render documents into high-quality PDFs at $100,000 an hour or more. How can having a PDF be compared to sitting in a plane several thousand feet in the air? When it comes down to it, a lot is at stake if documents can’t be rendered as required:

  • A Life Sciences organization must delay its submission to the FDA when the PDF document is not created, costing them up to $1 million in revenue per day
  • A Financial Services institution loses business to its competitors when they can’t produce insurance quotes in PDFs when the system is down or has a slower response time than its competitors
  • A Government agency loses personnel time when PDFs cannot be properly rendered to fulfill public service requests under the Freedom of Information Act (FOIA) or the Access to Information and Privacy (ATIP) regulation

Eliminating single points of failure

It is because of critical business reasons like these that Adlib supports high availability in our Advanced Rendering solution. Let’s face it, the IT infrastructures in large organizations are getting more and more complex, with multiple moving and interdependent parts. This can result in multiple points of failure.

By eliminating all single points of failure, it’s possible to minimize unscheduled system downtown, thereby maintaining or increasing the productivity in an enterprise.

Adlib achieves high availability by supporting redundancy for all software components in the system. However, it’s not advisable to have a redundant node idle until a failover occurs because of the costs of maintaining the hardware, power supply and real estate. Adlib counteracts this by implanting a support for redundancy using an Active-Active deployment, so that all nodes are always active and share the load. If one node fails, then the system still provides the service.

Achieving high availability

The Adlib components send a signal to the system database – kind of like a heartbeat – and components that share the same function monitor each other’s heartbeats. When one heartbeat stops, another component takes over the task within a few seconds. Some of our customers have told us that if IT doesn’t respond to an end user’s service request within 3 minutes, there will be a follow up, many times in the form of an angry phone call. This can all be avoided with high availability software. Alerts can be configured to notify IT once a component failure occurs, and they can take action to bring the system up to full capacity.

Delivering a high availability solution isn’t easy. It is about more than just quality code, system stability and product robustness. External factors such as hardware failure and environmental disasters also need to be considered. And while load balancing is important, it just helps to distribute work, but doesn’t guarantee that work isn’t lost. Delivering a high availability solution means that you have to deploy software that can be architected to detect and eliminate all single points of failure. Components work as a team, and if one fails, the others have its back.

Optimizing the mean time to repair

The mean time to repair (MRRT) is also an important factor to consider when deploying enterprise-wide software such as Advanced Rendering. Typically, critical business systems require 99.99% availability, which results in only 52 minutes of allowable downtime a year. Adlib has made this recovery easier by storing all of the configuration data in the system database. Provisioning a new node is as simple as installing the component software and defining the database connection for it.

If you’re interested in more information on Adlib’s Advanced Rendering products and how we maintain high availability, check out our product guide. Luckily, planes often have multiple engines, not just to balance the work but as redundant power sources if one ever gives out.

Next week I’ll be discussing the importance of easy integration capabilities within an enterprise. Stay tuned! If you missed last week’s post on linear scalability, check it out here.

About the Author

Jeff Brand