High Availability and Fault Tolerance Part One

In computing environments we generally accept that there are many conditions that can result in unexpected downtime. Unexpected downtime is any condition that results in not having access to your systems, applications, and data. A power outage that causes your computer to shut down is a simple example of “unexpected downtime.” In computing environments we work diligently to prevent this condition through various facilities. To prevent downtime due to a power outage we use “Uninterruptable Power Supplies” (UPS) which is a nice name for a battery pack designed to be on standby and “stand in” to keep your systems running until the power comes back online. We also can take the next step and put standby generators in place that will run indefinitely until the power is restored. The safeguards can be regarded as steps to provide “High Availability” (HA) and “Fault Tolerance” (FT) for computing systems. Over the next few weeks I will be publishing a series of blog posts exploring the difference between HA and FT in various environments and differing levels of complexity all the way up to how we use Stratus Technologies everRUN MX to provide guaranteed “five nines” of uptime for windows server workloads. A guarantee of five nines of uptime means that you should experience less than 5.25 minutes of unexpected downtime per year. We have worked with the everRun technology for over 15 years and our experience with this product has been that it performs as advertised and solves the uptime issue where many others fail. In addition to being able to protect your systems from server failures it is also possible to protect an application across multiple datacenters over distance. Besides its amazing track record everRun also allows up to 8 vCPU cores which is an amazing feat that no-one else in the industry can offer. Please stay tuned to our blog for more information on this exciting technology.