A year and some months after ESnet was honored with the Excellence.gov award, it is time to reflect on how our organization is tackling the challenges of running a production network while supporting major projects like the Advanced Networking Initiative.
Last year we added nine new people to the ESnet team: Hing Chow, Andy Lake, Chris Tracy, Josef Grosch, Inder Monga, Greg Bell, Sowmya Balasubramanian, Wendy Tsabba, and Steven Chan – some in part time roles. The challenge has been to maintain organizational excellence while scaling up.
It has been a month now after our recent reorganization, time to re-examine motivations and evaluate early results. Prior to this, ESnet was organized into separate, well-defined teams, each responsible for their area of expertise. The Infrastructure group managed the systems supporting ESnet’s internal business processes. The Network Engineering group handled the design and day-to-day operations of the network. The recently created Advanced Network Technologies group had a clear mission to conduct network research to develop new capabilities and services tailored to meet the needs of the research community. This structure worked efficiently as long as teams worked within their own domain of expertise. As the projects became more complex, a gap appeared in ownership of getting these components to work together. The old model of network engineers communicating their users’ needs to programmers who developed the tools and the system administrators who supported them resulted in slow and cumbersome integration. The ‘systems approach’ was often discussed but rarely followed.
But emerging technologies, virtualization and converged infrastructures are beginning to blur the lines between the traditional roles of R&E networking and computing. It recently became clear to me that if ESnet was to deliver high-performance, end-to-end solutions rather than point technologies, we needed to adapt the organization to a new paradigm. The siloed approach had ceased to be fully responsive to the call for seamlessly integrated storage, network and computing. Upon closer examination we realized our network engineers were already effectively writing code, system admins were becoming a lot more familiar with networking, and operational teams were tackling research topics. It was time to formalize a more effective way of working.
The July 1st reorganization turned ESnet into a flatter organization with a greater emphasis on teamwork. Greg Bell is the new Area Lead for Infrastructure and Support. His primary responsibility is to ensure a consistent approach across teams toward building end-to-end solutions. He’ll be working closely with Inder Monga, Area Lead for Research and Services. Together, they’ll ensure project teams are formed with resources drawn from the different skill sets across the organization. A team’s work is no longer complete when it is handed off to the next group, now at the successful conclusion of the project. While the reorganization did not dramatically change our organizational structure or roles of the existing teams, I believe it resulted in a change in mindset.
We are continuing to grow – so if you are looking for a challenge, are a network/software engineer and are interested in enabling big science through networking, then send your resume to me at firstname.lastname@example.org We are looking to add new stars to this excellent team that I have the pleasure of working with every day.