The last several newsletters have focused heavily on a look backward and forward. In this process, we’ve put forward our long-range predictions that the future of corporate IT will be “in the Cloud” and accesses by fixed and mobile devices of all types. We also postulated that security will continue to increase in importance.
The last several newsletters have focused heavily on a look backward and forward. In this process, we’ve put forward our long-range predictions that the future of corporate IT will be “in the cloud” and accesses by fixed and mobile devices of all types. We also postulated that security will continue to increase in importance.
In this newsletter, we will look at two remaining issues: Network Performance and Legislation. And though it may not be apparent at first, we believe that they will turn out to be integrally linked.
First, performance. For the most part, application performance in many aspects has been pretty-much optimized for corporate networks that are not in the cloud. (And, by the way, we can’t leave this review without pointing out that for the history of computing and communications, a “cloud” has always been drawn to represent an abstraction of “stuff happens here but it’s too complex to draw.”)
But now cloud performance is emerging as a major issue, and it is orders of magnitude more difficult than optimizing performance for a typical corporate network. In particular, the traditional corporate network has a finite number of communication links with a finite number of well-known processors/servers. But with the adoption of cloud networking and virtualization, this is now a multi-variable and extremely complex web. If there’s any good news, it’s that we’ve dealt with these uncertainties over the years on a much less grand scale, for instance as we moved call traffic engineering from a traditional telephony to a VoIP infrastructure.
Nevertheless, this will be a major challenge, and it may well be a part of the establishment, as we mentioned in the last newsletter, of a parallel “Commercial Internet” to augment and complement the “Public Internet.”
Which is where we get to legislation.
Historically, the ability of regulators to keep up with current technology has always lagged by at least five if not 10 years. (For instance, the 1984 rules concerning “protocol conversion” in the network haunted carriers who wanted to offer Frame Relay to ATM interworking 15 to 20 years later.) And this is just the U.S. view. The worldwide gap between technology and regulations covering that technology will only continue to haunt and frustrate us.
Nowhere is this more evident than in the current net neutrality discussions. As we mentioned a few weeks ago, this discussion could potentially have even more severe implications for corporate networks than for public networks.
One view would be that service providers can provide differentiated services, but they must divest themselves of various business units so there is “equal access.” (Wait a minute – Are we back to Orwell and 1984? Literally?) But this is not too likely in the U.S. at this point. More likely, we will see strategic business alliances bolstered by a legislative laissez-faire attitude.
We do have a final prediction that we will stand behind 100%. The next 10 years won’t be boring.
