Virtualization Technology News and Information
Article
RSS
Getting into Virtualization with VoIP

Article Written by Matthew Hilton

As a technologist, you want the most advanced solutions that are not only fast and efficient but that will also sail through any committee-based approval process. In reality, most of your projects will be met with scrutiny and likely fail some degree of what I refer to as the "Too Test" - it's "too expensive," "too confusing and difficult," or "too time consuming." You've likely had a project stopped-cold because a decision-maker, usually from the management team, finance department or a consultant, decided the technology you proposed was "too" (fill-in-the-blank). Two major projects typically challenged with the "too test" are virtualization and voice. 

There is a silver lining. Whether you're looking to virtualize your entire infrastructure, or simply add a VoIP/unified communications solution to an existing virtual environment, with the right approach, these projects also usually ace the "too test." To combat what might seem like constant pushback, there's a simple process to help ensure a successful deployment with minimal downtime, less cost and higher customer success rates.

It all starts with a plan

Before looking at quotes, vendors and software packages, you need to develop a detailed plan of attack. Sure, "making a plan" sounds obvious, but it's easy to overlook. When approaching a virtualization or voice project, here's a basic four-step process to implement:

1) Assess the current environment

No matter how accurate the company's documentation is, and no matter how often you step into your server room or data closet, I'd be willing to bet there are some unknowns hidden away in there. So, first things first. Take a good hard look and create (or verify) your current environment. This should be more than just OS and function. Document the make, model, serial number, OS, function(s), and every piece of information you know (or can retrieve) for each and every piece of equipment in your server or data closet(s). Especially if you come across old devices that have been powered off or are unknown, document those as well. During this process, it's also helpful to find out how hard each device is working. Do you have servers that are running only 10-20 percent CPU usage consistently? Or do you have servers that could use memory upgrades? Make note of these answers too.

During this assessment, it's important to also note your human resources as well. Do you have a team of workers? Do these workers already have specializations in server, virtualization or VoIP areas? Or, are you a one-man shop? Do you utilize consultants or VARs? Gathering all of this information now becomes a key resource document in moving forward.

2) Strategic Evaluations and Timeline

Here is where the fun really begins. With your existing environment mapped out, it's time to start evaluating what your overall virtualization platform will look like in the future. All of the major players have multiple different versions of their solutions, from Citrix, Oracle, Microsoft to VMware, and they all come with varying options and price points. And this is where starting a new virtualization project with voice really can help. For example, Digium's Switchvox Unified Communications platform comes pre-packaged as a single OVA file. So, all of the difficult work of setting up your VMware guest OS is done for you and the voice platform is already optimized for VMware's unique tools and optimizations. So, half your work is done for you and it allows you to dip your feet into the virtualization pool, slowly and steadily. And as you progress and your knowledge on the platform (VMware) continues, you can grow and scale out that solution in your own time and with your own comfort level.

During this process step you should also map out suggested dates and timelines for your project implementations. If you need to address network concerns before moving to virtualization or voice, map out and time the purchase, configuration, installation and support time it will take before you move on to the next phase of the project. I like to also add in some alternate times here as well. If the network cutover doesn't go as expected, rather than rush my way through and "just get it done," I set about defining my rollback strategy, or my post-mortem/re-assessment and reconfiguration time; and, then, my second-cutover attempt timeline is in this plan as well. Ultimately, this will help lower the burden and stress on all staff and also minimize potential downtime to critical infrastructure needs.

3) Address the "toos"

Armed with a solid process and an assessment of your existing environment, many of the "too" arguments start to fall away without the need for complex counterpoints or concessions. Take the "too difficult" argument. By utilizing a virtualization platform that matches your needs, without over-indulging, and potentially even rolling out a free-version, the difficulty of the total solution can be minimized. As part of a rolling plan, that includes future upgrades and paths to upgrade to more feature-rich versions, the complexity of the entire project can be spread out over a longer duration making up front challenges more easily managed. In the long term, your entire plan may encompass very large, expensive, and complex back-end servers, virtual networks, shared storage and even advanced features and functions to increase uptime and automated recovery. In the short term, a single-server virtualized local-host may be all that's needed to prove the concept that virtualization does not have to be overly complex.

Similarly, each of the elements of the "too test" that I've listed (and even those that I haven't, but you're sure to run up against) are likely to be easily contested and resolved by implementing a slow and steady rollout approach. Starting small and working towards a larger infrastructure not only defers much of the stress and complexity, but it also allows your team and your entire organization to become familiar with and work on any bugs in the solution, before it becomes a critical business system. Resolving an early deployment issue while the solution(s) are still in test or pre-production or even early production phases are much more manageable than rapidly working to address a production-down situation.

4) Aim for value, not cost

Regardless of the choices you make and the rollout strategy you choose, there will be costs associated with moving forward. And it becomes quite easy to equate cost and value. But this is where the benefits of virtualization really shine. Being able to stand up multiple virtual servers, appliances and applications onto a single virtualized hardware platform, means that cost over the long run can be significantly reduced. And because your infrastructure becomes virtual instead of physically based, upgrading to a larger scale becomes more cost effective too, as older hardware can still be useful and not just powered off or thrown away.

The overall picture

As unified communications solutions continue to advance beyond voice-only phone systems, their role in critical IT infrastructure continues to expand. And as it does so, the necessity for these solutions to remain operational with high levels of redundancy, availability and stability become even more valued. Standalone hardware appliances and legacy phone systems with proprietary hardware are looking less appealing to current buyers. And concurrently, while businesses are continuing to stretch their bottom lines, spend less on IT infrastructure yet demand high-end services, virtualization is continuing its march towards dominance over standalone, hardware-based, one-to-one physical servers. By combining virtualization technologies with a future VoIP phone system rollout, your company benefits from the values that both solutions provide, all while minimizing downtime and cost.

##

About the Author

Matt Hilton 

Matthew Hilton is a product marketing manager at Digium, a business communications company based in Huntsville, Ala., that delivers enterprise-class Unified Communications.

Published Friday, December 01, 2017 8:04 AM by David Marshall
Filed under: ,
Comments
There are no comments for this post.
To post a comment, you must be a registered user. Registration is free and easy! Sign up now!
Calendar
<December 2017>
SuMoTuWeThFrSa
262728293012
3456789
10111213141516
17181920212223
24252627282930
31123456