I am the first to admit it. I don't like going to either the doctor or the dentist. It makes me uncomfortable when someone stares into my ears, up my nose, or into my mouth. I don't like being poked, prodded, and told to "turn your head and cough."
However, I do recognize the value of catching problems early in their lifecycle. I know far too many people who waited to see a healthcare professional only to find out that their nagging pain in the stomach was a very serious matter. So, despite my discomfort with blood pressure cuffs and tongue depressors, I understand and appreciate preventive medicine. It's much better to suffer minor discomfort now than deal with a life altering issue after it's too late.
While the health of your communications system may not be as important as the health of your body, it too needs regular checkups and potentially, adjustments. Like your little aches and pains, temporary lapses in service or poor voice quality can be signs of something much bigger. Ignoring a problem, hoping it will go away on its own isn't a good strategy ... whether we're talking about humans, software, or machines.
Like the human body, your communications system has a lot of moving parts. At the core is some form of call processing engine. This is the server that manages users, features, access elements, and peripherals. Surrounding the core are adjunct servers that provide everything from voice mail to contact center functionality. Access elements, such as trunks, connect your system to the outside world. Soft, hard, and mobile endpoints provide unified communications to your users. Finally, management tools allow administrators to configure all these discreet components into a cohesive whole.
With the advent of IP communications, it would be foolish to ignore the underlying network that connects all these pieces together. This includes routers, switches, wireless controllers, and even the cabling that runs through floors, ceilings, and walls.
Every one of these pieces plays an important role in delivering communications in and out of an enterprise, and the tiniest failure can sometimes create major headaches. One bad or improperly configured port on a data switch can be all it takes to bring your voice traffic to a grinding halt.
My experience has been that a communications system runs at its peak performance around six months after installation. All the initial integration glitches have been worked out, and the configuration is as pure as it will ever be.
As time goes by, however, the flaws start to appear. New servers are added, and new services are fired up. Your once pure configuration has been muddied up by different administrators, each with his or her own unique ideas on maintaining and enhancing the system. Software and firmware become out-of-date, and components are upgraded independently of each other. New trunks and endpoints are added. Contact center managers add additional call routing options. Network components are swapped in and out without consulting the telecom department.
That previously well-oiled machine now spurts and sputters. Voice quality suffers from noise, echo, and worst of all, one-way audio. Calls drop unexpectedly or won't connect in the first place.
I sat down and thought about all the places that can start out right, but over time, work their way to wrong. This includes hardware, software, configuration, and practices.
The easiest way that hardware can go wrong is for it to go out-of-date. This could simply mean there is a replacement that runs faster and can handle greater capacities, or it could be older than dirt (which is about five years in our industry) and no longer supported.
Slower processors might be fine if you run the necessary calculations and discover that you don't need the latest and greatest. That's similar to me and my iPhone 5. The iPhone 6 has been around for a year or so, and I still don't see a reason to upgrade. I never recommend change for the sake of change.
Manufacture discontinue is a much more serious issue. While it may be possible to find spares on the gray market, you are running a serious risk of downtime when you continue to use unsupported hardware. Sadly, I see this far more often than I would like. We are used to the long mean time between failure (MTBF) ratings of traditional communications equipment and grow far too complacent.
There is also the case of too little. I commonly run across this with Digital Signal Processors (DSPs). While the capabilities of DSPs are quite extensive, here in the communications world they are mostly used to convert between forms of media. For example, a DSP can convert G.711 to G.729 and vice versa. More commonly, DSPs sit between IP and TDM calls to enable communication between these disparate technologies.
Too many DSPs is a waste of money and too few create configuration and runtime problems (no ringing, dead air, etc.).
Long gone are the days when you can say your communications system is running software version such-and-such. That's because a modern day communications system is comprised of many different moving parts. Your call processor might be running the latest and greatest software, but the firmware on your ISDN cards could be several versions back. The same might be true about your voice mail system or the firmware on your IP telephones. By decoupling services from the core, we've lost the ability to apply an upgrade to one place and be done with it.
Like out-of-date hardware, old software can cause all sorts of problems. In addition to missing important new features, mismatched software versions can create problems when you connect different components together. These problems might be obvious (i.e. they won't connect) or show up as unexplained issues every so often.
Dated software creates problems with supportability. Manufacturers typically stop producing patches for old software and some of those patches (e.g. security fixes) might be absolutely essential.
New software might also force you to rethink system configuration. For example, the earliest versions of Avaya Aura supported IP Network Regions in a limited fashion. Newer versions continue to support the original data structures while adding region types that didn't previously exist. So, while the new software is backwards compatible with the old, administrators that don't go back and reconfigure their IP Network Regions miss out on current best practices methodologies.
Lastly, old software prevents you from taking advantage of exciting technologies such as Communications-Enabled Business Processes (CEBP) and the Internet of Things (IoT). The interfaces and features that enable these new platforms didn't exist until very recently.
Earlier on in my career, I would regularly sit down with contact center users and supervisors to analyze workflows and procedures. It was amazing how much I was able to uncover about a company's contact handling of which no one in the company seemed to be aware.
Hardware and software are important, but just as important are how they are being used. What might have been effective years ago may be antiquated and counterproductive today. Contact interactions are evolving, and it's essential that the ways your customers are serviced keeps pace with these changes.
Periodic health checks that look at all the aspects I described above are key to a robust and long communications life. It's essential that this holistic approach be taken to ensure that you understand the state of each major subsystem as well as how each interacts with the others. It's not enough to look at hardware while ignoring software. On the flipside, new software likes newer hardware.
Unfortunately, it has been my experience that many IT departments can't tell you the state of all the piece-parts that make up their communications systems. They aren't aware of the many levels of firmware that exist on their line cards and IP telephones. They are unsure if they've employed too many or too few DSPs. They can't tell you if their configurations follow best practices or are a hodgepodge of 10 years of changes by 30 different administrators.
With the exception of a workflow audit, it's certainly possible to do this work in a manual fashion, but it's not advised. There are automated tools that interrogate every aspect of a communications system to produce detailed reports. For a complete and thorough view, you need deep dives into every moving part.
Similar to a financial audit, I highly recommend that you engage an outside company to perform your checkups. Due to a variety of reasons, the owners of large communications systems are often not the best people to evaluate it. There tends to be too much "see no evil, hear no evil, speak no evil." We see what we want to see and ignore the ugly parts.
The end result of a good health check is simple. You know what you have, know what's wrong, and know what's required to make things right again.
Now, open up and say "Ahh."
Andrew Prokop writes about all things unified communications on his popular blog, SIP Adventures.
