Managing UC – you need to be in control
No room for backseat drivers
Workshop Love it or hate it (and from the feedback we have seen from Reg readers, the jury is still very much out on this one), Unified Communications (UC) is here to stay, and it will increasingly make its presence felt in the workplace. Not only will it change the way people work, but also the way the IT environment is managed.
While it's important to develop a new approach to working, to really see the benefits from UC, it's even more critical to ensure that UC is properly managed from a technical perspective. After all, a service that isn’t being properly used won’t necessarily make a business grind to a halt, but a service that is fraught with technical problems just might.
An unwelcome side effect is that it will inhibit its take-up by staff as a whole. Bad experiences can negatively impact service credibility, and IT along with it. Certainly the old truism ‘first impressions last’ is one worth remembering here.
At the core of a UC implementation is the network, so making sure that this is up to the job will be time well spent. This will involve looking at how current connections are utilised in various locations – eg home offices, branch offices and central offices – and what might be required in a UC world.
An example of this might be a branch office which has sufficient bandwidth for day to day use for normal communications, as well as allowing for periodic data backups to central servers. But if the company calls an ad-hoc video conference meeting that pulls in several departments based in that branch office, will the network be able to handle this from a capacity and quality of service management perspective, particularly if it happens to coincide with the scheduled backup? If all it can support is a mediocre performance, when it has been billed in the top slot, its proponents and architects will be left squirming in their seats.
More importantly, the perception of what UC can deliver will suffer a blow that it just might not recover from. And while video conferencing isn’t the only example we can draw on, it is a useful one. Video is an application that isn’t particularly fond of latency and low bandwidth. To quote one reader: "Make sure your bandwidth is up for it. Especially when looking at video solutions."
The point that this raises is that UC will force the IT department to approach management of the network and of applications in a very different way. Carrying on business as usual is not an option. Over recent years, the need to adopt a very proactive approach to network and application management has diminished to some degree, thanks to network connectivity working so much better than in the past. Networks are a lot more stable than they used to be, with much better quality of service (QoS) - as are business applications, in the main.
With a UC overlay, and the potential for scenarios such as the one described above, where an inappropriately scheduled video conference could topple the network, the need to take on a much more active role with respect to monitoring and management becomes evident. This is particularly true where UC is interfacing with business-critical links or applications. And while video may be a rising star of UC, voice is anything but. Like video, voice communications need good networking. No one likes to have dropouts and unintelligible conversations on the phone as a result of sub-optimal connections.
As a consequence, ongoing network monitoring and active management becomes essential, not just in the early weeks or months. With UC, the likelihood is that usage will ramp up as more people become familiar with it and incorporate it into their working patterns. And because of this, the days of getting by on a wish and a prayer for IT provisioning are probably well and truly over. The need for tools to measure performance of different aspects of UC such as voice or video quality become much more important, especially from remote or less central locations. As one Reg reader observed about getting UC running in the business:
It is difficult -- networking, video, authentication and authorization, scheduling, integration with business processes and staffing all are tough nuts to crack.
While all companies are pretty much in the same boat in this regard, it is probably the mid-sized companies that will be most affected, as they are less likely to have networking and applications specialists on the payroll.
Where services are being delivered via the managed or hosted route, the expectation is that the vendor or service provider would have adequate tools in place to carry out thorough quality of service monitoring, but this shouldn’t be left to chance. There is little point in complaining that the ‘agreed’ service level agreement (SLA) falls short of company requirements once something goes wrong, if it wasn’t properly negotiated in the first place. In short, you should aim for an SLA that reflects your own needs as closely as possible, and agree upfront just how the SLAs will be measured and reported.
At the end of the day, if you are happy to take more of a back seat when it comes to managing UC across your business, don’t be surprised if you don’t end up where you really want or need to be.
What nuggets of technical wisdom can you impart around managing UC? Have you seen an implementation go off track because of technical nightmares? Or do you think you have best practice down to a tee? Do share. ®