Networking

12:41 PM
Commentary
Commentary
Commentary
50%
50%

A Plug For Plugfests

Standards are great, but there can be significant differences in vendor implementations. The University of New Hampshire's InterOperability Lab provides a neutral venue for testing and analysis.

Having spent a large portion of my career working in industry for a variety of companies, managing consortiums at the University of New Hampshire’s InterOperability Lab (UNH-IOL) has given me tremendous perspective on how technology is implemented in the data communications industry and the value of a "plugfest."

The UNH-IOL provides independent, broad-based interoperability and standards compliance testing for data, telecommunications, and storage networking products and technologies. The UNH-IOL consists of roughly 25 different year-round standards-based testing programs called consortiums. Each of these consortiums represents a collaboration of industry forums, service providers, test equipment vendors and otherwise competing companies. In addition, the UNH-IOL also hosts plugfests -- multi-vendor group tests -- as often as four times a month.

The bulk of my career was spent working for data communications equipment manufacturers. My career started as a firmware development engineer and evolved into manager of a system test team. In each of these roles, I was constantly and consistently made aware of how far behind we were from a feature perspective. New industry standards were released often and we were barely implementing features that were important lastyear. Our roadmap was long and robust -- but late. In system testing, our schedules were often compressed due to slips in development and we spent long hours brainstorming ways to solve the QA bottleneck because again, we were late to market.

Having been at UNH-IOL for three years now, I have learned that many companies experience similar delays. Implementation of new industry standards can be an arduous task that has to be accomplished while maintaining current products in an installed customer base. Engineers at companies are often shy about saying their implementation is ready while the marketing and sales teams say it's good to go. If you are the industry leader in a certain feature, it’s difficult to prove your implementation if other players in the space are slow to implement. This is where UNH-IOL has a role to play.

In the plugfests at UNH-IOL, companies come into the lab for a week's worth of testing on a pre-approved, often collaborative, test plan. The entire event is under NDA and the purpose is to create a neutral environment for the betterment of the products, as well as meeting industry standards. Engineers from competing companies solve complex implementation problems at these events, essentially helping each other and creating a better data communications product.

I have had the pleasure of running two plugfests, both for Transparent Interconnection of Lots of Links (TRILL). For those that don't know, TRILL is a solution to an outstanding Spanning Tree issue. It competes with proprietary technologies to the Spanning Tree problem, as well as other standards, such as Shortest Path Bridging (SPB) and OpenFlow. Because there are competing technologies in this space, many companies are slow to implement, waiting to see which standard will take hold. Others are implementing more than one of the technologies, which in turn slows development because their development teams are spread thin. For the company that is implementing to the TRILL standard, it has proven to be a somewhat daunting task to prove their implementation due to these factors.

[TRILL can help with the Spanning Tree problem but what will its role be in software defined networks? Ethan Banks examines the issue in "Will SDN Kill TRILL?"

Our most recent TRILL plugfest took place in May. Extreme Networks, HP, Huawei Technologies and Ixia participated. Our test methodologies were designed to create a multi-vendor TRILL campus to prove that each vendor’s implementation would function in the campus and communicate properly to each other using the TRILL protocol. During the event, there were active discussions among these vendors as they tried to better understand and interpret the TRILL standard. We were able to collectively troubleshoot and bring up the network. The engineering teams involved in the plugfest gained valuable insight into where their product was in terms of implementation. They left with a stronger understanding of TRILL and of what is expected of their product.

I look back to my development and QA roles and think about how much I could have learned from participating in plugfests. I feel strongly that taking part in a plugfest would have greatly expedited time to market. In addition to the value for the development process, companies that are interested in deploying a new product or technology can find solace in knowing the product or technology they are about to deploy has actually worked with other vendors' products in that space.

Christina Dube is a senior manager and acts as the HR lead for the executive committee at the University of New Hampshire InterOperability Laboratory (UNH-IOL). She is responsible for management of the Bridge Functions, Data Center Bridging and Fibre Channel consortia.

Comment  | 
Print  | 
More Insights
Cartoon
Slideshows
Audio Interviews
Archived Audio Interviews
Jeremy Schulman, founder of Schprockits, a network automation startup operating in stealth mode, joins us to explore whether networking professionals all need to learn programming in order to remain employed.
White Papers
Register for Network Computing Newsletters
Current Issue
2014 Private Cloud Survey
2014 Private Cloud Survey
Respondents are on a roll: 53% brought their private clouds from concept to production in less than one year, and 60% ­extend their clouds across multiple datacenters. But expertise is scarce, with 51% saying acquiring skilled employees is a roadblock.
Video
Twitter Feed