Scalability/ Redundancy Over the years, we have many times been approached to take over an existing digital signage network that may have functioned acceptably during a small pilot project or early on in smaller numbers, but once the project expanded, things rapidly degraded. This can occur for a number of reasons. Software applications and the infrastructure they are built upon are generally designed to handle certain volumes of traffic, once you exceed these volumes, things get congested and often fail. That is why scalability is important.
Consider a country road for example. Under circumstances for which it was designed, it works perfectly. It’s cost effective and provides access to more remote areas. Now imagine increasing traffic on this country road so as to snarl it with heavy traffic and possibly some inclement weather; it would quickly cease being a viable roadway. On the other hand, major roadways like Enterprise Class software infrastructure, are designed to support everything from a Sunday drive to snow covered holiday traffic. It’s all in the design and the infrastructure it's built upon.
Ensuring High Availability is more than just bandwidth, it’s also about the redundancy built into the infrastructure; all of the hardware, software and connectivity. This is especially so when considering SaaS based or cloud based software systems. A web based software application running on a single computer server connected to the internet, sitting on a rack in a wiring closet, can technically be referred to as a SaaS Based cloud offering. Using our roadway example from above, it may work well during a small test, but if it gets hit with higher volumes of traffic or any “inclement weather”; say a hardware failure or a lengthy power/Internet access outage, the system will cease being viable. On the other hand, true Enterprise Class SaaS based solutions are designed for High Availability rates of at least four nines of uptime (99.99%).
Redundancy is an important part of this process. Ensuring that not only Internet access but also instruction servers, database servers, file servers, and web servers are all properly load balanced and redundant for maximum possible availability is essential. In a High Availability Enterprise Class software solution implementation, there can be no single point of failure. To that effect, protection against catastrophic failure must also be considered. To that end, Enterprise Class software companies like us go to the extent of keeping a redundant, high availability infrastructure on “Pilot Light” on an opposite coast. Not an event that is pleasant to think about, but for completeness, one that needs to be there just in case.