Applications are hot, and UC/CEBP are the driving forces that will, as our upcoming VoiceCon virtual event is titled (stay tuned!), "Transform the Enterprise." But I want to take a minute here to argue for the importance of network infrastructure and the management thereof. (For an additional take on this topic, see Irwin Lazar's blog on Enterprise 2.0.)The recent groundswell of interest around SIP trunking brought this home again to me. In our massively-attended webinar last week (replay here), the question came up about the new burden that SIP trunking places on network managers to ensure that the appropriate network assessments are done and that the proper QOS is enabled to support real-time traffic over SIP trunks.
Networking hasn't been sexy for awhile now, but I don't think it's ever been more important that it be done right. We've known this in the communications world ever since we first started trying to run voice over IP networks with, at first, mixed results (to be generous).
If you follow John Bartlett's blogs on this site, you get a sense of some of the challenges that confront enterprise professionals who must support demanding, bandwidth-consuming real-time applications on IP networks. We're moving into a world where video and other heavy-duty applications are constraining basic bandwidth capabilities, while at the same time network latency is an issue thanks to distributed networks (deployed in hopes of saving money, being green, and creating more resilient networks in the face of disaster, disease, etc.).
In the blog I linked to above, Irwin mentions software-as-a-service (SaaS) and cloud computing as two trends that are helping to drive the industry toward a fresh look at the Internet's architecture. I'd argue that these trends, insofar as they represent the possibility of blending premises-based and cloud-based functions, only add to the need for enterprises to make network performance their top priority.
To the extent that SaaS and cloud computing begin to play a role in enterprise communications, they'll require a high degree of coordination across the public-private network boundary, as well as between diverse public networks that may exchange traffic. Right now there's a mind-boggling diversity of QoS handling, codec use, and reporting and management among these various networks, suggesting enterprise network managers will have a difficult time taking full advantage of the new technologies while at the same time being assured that their traffic is being handled satisfactorily.
Cool applications and business process integrations are another whole challenge, one that's organizational as well as technical. But the better your team is able to solve the application payoff challenge, the more critical it will be that those applications can actually perform as planned, so that you actually capture the benefit.
