Aug 11, 2016

Critical Systems? You need High Availability

More than ever, modern technologies are a critical part of the world around us. Integrated into business, entertainment,health and countless other applications, anything that impacts the availability of technology services can affect work productivity, affect the bottom line, affect health, or cause major inconveniences. When it comes to business, there is little doubt that the result of a system downtime can be measured in terms of lost opportunity, lost revenue, lost reputation, and lost productivity. Those are the reasons why we talk about High Availability – risk, impact, and costs. We need systems to run and be available at all times.

How critical are your systems?High Availability

If you think a system might be important, the task of finding risk and assigning critical importance generally comes down to a few questions:

  • Does the information system in question generate revenue?
  • How significant would the lost revenue be from a system that’s down?
  • Could a down system result in lost customers?
  • How would downtime affect the productivity of employees?
  • Can your business afford a loss of trust and reputation if critical systems are down? What is that threshold?
  • Is the system in question “mission critical”?

Not exactly science, but it’s a worthy evaluation to conduct. After evaluating various risks and potential costs, the critical nature of a business system emerges. The next step is planning to avert downtime and High Availability is the best weapon in this effort.

Explore High Availability

HA means that infrastructure is designed and deployed with redundancy and the ability to “failover” automatically. Essentially, minimizing or stopping infrastructure downtime through infrastructure itself.

It’s a bit of science and a bit of principle, but by eliminating single points of failure at various levels, one can increasingly address the constant availability of a system. HA elements include load balancers, firewalls, shared storage, web servers, application servers, and even databases deployed in a redundant configuration that is able to failover in the event of a failure or planned maintenance operation.

If HA configuration is implemented at each of those levels, Full High Availability is achieved.

  • Load balancers can be configured in a redundant setup. If one of the load balancers fail for any reason, the second steps up to handle all of the traffic management and delivery.
  • Firewalls can be configured in a similar manner. The loss of a primary firewall initiates a cutover to a secondary firewall so that production is not affected.
  • Database servers can be configured to automatically switch primary control from one server to the other
  • Storage systems can be configured for simultaneous system access in a shared configuration
  • Web and Application servers serve as mirrored instances of the “front end” of an application delivery

In the most critical of deployments, a system will see all of those elements fully configured in HA design. No chances are taken.High Availability

Nonstop, everlasting computing

One of the advantages of High Availability is that you never have to stop the entire system. That goes for those times when you have things to fix, especially if something fails, and it also covers those times when you have to expand components within the system itself.

Every system requires some sort of maintenance, some sort of patching, or some sort of upgrade from time to time. HA is designed to allow for planned downtime of components for this exact purpose. For example, you can patch or fix one system while the other runs, then turn it back and perform the same on the other system. Along that same track, when it comes time to upgrades and growing the infrastructure, the same HA construct makes these operations easy.

Critical needs meet specialized solutions

High Availability is one of the best insurance policies against unwanted systems-based losses. However, it’s complex and may not be realistic for everyone. It requires specialized design, experience, maintenance, and regular testing. Codero’s Managed Hosting solutions can make that easy to deal with high volume, application requirements, and most importantly, critical business needs. Chat today with one of our Codero experts.

Share on FacebookTweet about this on TwitterShare on Google+Share on LinkedIn