CloudE 1.0 Aims to Be Ready Before Its Time

Yesterday, I wrote about the changes in how technology is commercialized by vendors and service providers. Two small events — the recall of batteries by Panasonic and a potentially dangerous design flaw in the Nest Protect from Google — may or may not have actually been affected by the accelerating “velocity” of new services and apps. The idea is that things are rushed to market so quickly that the due diligence of years past is not done.

The evolution in how technology, services and applications come to market is front and center again today. Light Reading contributing editor Robert Clark details work by the CloudEthernet Forum (CEF) on the CloudE 1.0 spec. The approach is subtly different from the traditional way of setting standards. But it is a very important difference. Writes Clark:

Telecom industry groups typically take two to four years to create a standard, says CloudEthernet Forum president James Walker. But a cloud industry body working at that pace would render itself “irrelevant.”

The idea is for the CEF’s Open Cloud Project (OCP), in conjunction with the Metro Ethernet Forum, to establish what the organizations call a “reference architecture” that will enable providers and enterprises to match what they are doing against the current version of the architecture, even before it is promulgated. Instead of waiting until all the “Is” are dotted and “Ts” crossed, the idea is to enable the technology to move ahead in an incremental “so far, so good” fashion.

This isn’t the first time that such an approach has been taken. New iterations of 802.11 Wi-Fi specs, which seem to be released every few weeks, are available to vendors long before they are official. The thinking is that the technical community is savvy enough at this point to create a detailed vision of what the spec eventually will look like to be used without fear that the actual final version will be significantly different.

A complicating factor in the case CEF is that it is a cross-disciplinary undertaking. No Jitter’s Bob Emmerson indicated that it includes software-defined networks (SDN), network functions virtualization (NFV) and carrier Ethernet. Working across these lines will make the crafting of standards a particularly difficult challenge:

Several enabling technologies are needed to increase the agility and lower the cost of cloud computing infrastructures. Therefore, consensus and a shared vision on cloud Ethernet architectures and technology priorities are needed, and the various players must unite on open standards, interoperability and easy deployment of cloud services. In a nutshell, that is the key objective of the CEF.

This suggests a longer and more complex road to a specification, and makes the incremental process seem prudent. The CEF is clearly trying to get all the players on board. A CIOL story maps the close relationship the organization plans with its members. There will be an “open day” in late July and ongoing consultations through the summer.

The new approach of The Institute of Electrical and Electronics Engineers (IEEE) and CEF in the development of Wi-Fi and CloudE specs, respectively, speaks to the same issue: Things move too quickly and too much money is on the table to not proactively push the creation of standards and specifications.