Dodging a Digital Data Disaster
It’s been pretty firmly established: The Digital Revolution is inevitable. The Convergence Wave is coming crashing down whether we like it or not. That being said, there is still much debate on the merits and benefits of old-school analog systems.
While video quality (even I still believe good analog video on a good analog monitor looks superior) is a popular argument, the one I want to focus on is resiliency — the ability of a system to just run, and keep running. Remember the good old days? You never had to worry about viruses taking your multiplexer offline. You never had to protect your matrix switch from hackers. These devices just worked.
I’m not saying we need to take a step backwards. But we do need to look at some ways to build more resiliency and robustness into our new digital systems. Even today, some end users that have been comfortable with their existing system for many years are hesitant to jump into the new technologies.
Now that our systems are moving onto the network, we need to look at some common ways network designers and administrators not only protect the network and its content, but also attain the coveted 99.999999 percent uptime they all seek.
Redundancy Is Good
The first topic to look at is another ‘R’ word, redundancy. This is simply a method of duplicating layers into a network or system, whether it is duplicate hardware devices or network paths. Redundancy is simply a way to eliminate a single point of failure. It has been a failsafe option for a long time, even back in the analog days.
When it comes to the network, redundancy can be implemented at several points. At the core of the network, redundancy might require multiple network paths to neighboring switches, so that if one path fails, network data can easily be rerouted (or switched) around the problem link.
There are two main problems with this solution, however. One is the sheer number of links required. If you want full redundancy, basically a mesh network with all switches connected to all other switches, the calculation to figure out the number of links is n(n-1)/2. In other words, if you had four switches in your network, you would need a total of six links between all of them [4(4-1)/2] to achieve full redundancy.
To be perfectly honest, most of the time, a couple of redundant trunk links suffice. A full mesh generally isn’t necessary on LAN systems.
The other issue with this type of redundancy is traffic looping, otherwise known as a broadcast storm. If you give network data multiple paths to the same place, they’ll take them … all of them. What you have is switches forwarding traffic out multiple ports that all lead to the same place. This generates a great deal of traffic just floating around on the network, because the destination switch already received the first packet that came in, and is rejecting all the rest.
So the moral of the story is, as far as the network is concerned, even attempts at creating the desired resiliency, can result in problems.
Get a Handle on Hardware
Another way to build a resilient system is by using redundancy in your hardware. Any hardware component can be a point of failure. Some might be important enough to duplicate, in order to maintain uptime.
One component that can be duplicated in a server or storage array is the network interface. Much like the redundant network switch connections we discussed above, having multiple paths to the network on your server also serves as a protection against single component, or even network failure.
Multi-homing a computer, running two network cards simultaneously, has been used for a while in security projects to allow a single computer to access both a production business network as well as a video network, when the two aren’t otherwise connected. In this case, however, both network interface cards (NIC) would be connected to the same network (although not necessarily the same switch), providing a way to stay linked in case of a single path failure.
Even whole servers can be placed in a redundant situation. This is how large Web sites, especially e-commerce sites, stay up in the face of extreme loads or deliberate attack. A concept called load balancing is used, where each server communicates its current load to the other servers and if one is overloaded traffic can be redirected to other servers to compensate.
RAID Can Kill Data Loss Dead
The one aspect of redundancy and resiliency that has the most relevance to the security market is in data storage.
Say what you will about the long-gone VHS tape, but if a VCR was well maintained and the tapes were stored properly, there wasn’t much chance that the information on the tapes would just vanish. Sure, tapes could break, but it wasn’t like one little chuck of magnetic data on a tape could destroy the whole thing, or one tape failure out of a week’s worth of tapes could make the whole week disappear. Disk storage is susceptible to these types of catastrophic failures.
If you enjoyed this article and want to receive more valuable industry content like this, click here to sign up for our FREE digital newsletters!
Related Content
Security Is Our Business, Too
For professionals who recommend, buy and install all types of electronic security equipment, a free subscription to Commercial Integrator + Security Sales & Integration is like having a consultant on call. You’ll find an ideal balance of technology and business coverage, with installation tips and techniques for products and updates on how to add to your bottom line.
A FREE subscription to the top resource for security and integration industry will prove to be invaluable.
