Cluster computing is connecting a group of computers in a dedicated network so they can act like a single entity. The operating system can assign a unique number to each cluster and keep track of files according to which cluster(s) they use. There are a number of reasons for people to use cluster computers for business systems, from a need to extend the processing power of a single computer with the full computing capability of a cluster, to a desire to ensure that a computing system is always available. The latter form of clustering is called High Availability (HA) cluster computing.
High availability clustering is a method used to minimize downtime and provide continuous service when certain system components fail or need to be updated. HA clusters consist of multiple nodes that communicate and share information through dedicated real or virtual networks and are a great way to ensure high system availability, reliability, and scalability.
High availability is essential for any organizations interested in protecting their business against the risk of system outage, loss of transactional data, incomplete data, or message processing errors. These risks, in some situations, can mean millions of dollars in revenue loss, or in the case of video security installations, put assets or persons at risk. For a business interested in being fully operational and secure at all times, HA clustering is a practical solution.
However, HA systems entail an extra cost as server hardware needs to be duplicated, special redundant networks established and cluster-aware software deployed. The working principle in HA systems is that any component that stops working will just be bypassed, and a similar component will be switched in to ensure no or minimal outage.
Think of a tandem bicycle. If one of the riders for some reason stops pedaling, the other rider will ensure that the bicycle does not stop, and keeps on going. Before HA systems can be deployed, the cost of downtime has to be evaluated, since the failover process can take time. In general, the more redundancy built into the system, the lower the failover time, but higher the cost.
XProtect 2016 Advanced video management software can be deployed in HA clustering environments. However, careful planning is needed to ensure system integrity. We strongly advise to consult all relevant documentation at www.milestonesys.com/support/. The white papers on Milestone’s Support site are a prerequisite for the deployment of XProtect in virtual server environments. Customers with Milestone Care support services can also get valuable advice on clustering using this resource.
For additional information about this topic, check out the white paper from Milestone Partner Stratus Technologies called ‘Various Approaches to System Redundancy and High Availability’. This paper describes the technology behind HA clusters and the four major categories of HA systems, as well as the pros and cons with each approach.
by JOS Svendsen, Corporate Communications Manager, Milestone Systems