Interview with LISA'13 co-chairs Narayan Desai and Kent Skaar

A conference the size of LISA is no small effort. USENIX relies heavily on the contributions of the community, including the conference chairs. This year's chairs are Narayan Desai and Kent Skaar. Narayan focuses on building large-scale computational platforms and software at Argonne National Laboratory. Skaar works for VMWare, focused on distributed systems management and applications. I recently asked them about their experience co-chairing LISA '13.

Ben Cotton: Planning a conference at LISA's scale is not a quick process. When did you start working on LISA '13 and how much have you put into it?

Narayan Desai: We were approached to chair at LISA '11, and we've been thinking about this since then. Since we were the IT coordinators for LISA '12 (with Cory Lueninghoener), we didn't start actively working on the conference until the summer of '12.

Kent Skaar: We needed to finish up the talks for '12 before we could really focus on '13.

BC: What's one thing at LISA '13 you're really excited about?

KS: It would be our targeted efforts to raise the technical level of the program. Examples are John Looney's SRE tutorial, our Metrics workshop with Brendan Gregg, the data science angle with Hillary Mason, and Bruce Schneier talking about privacy issues. The underpinnings of this years conference is one of system engineering, which guided us to invite many of our speakers and also motivated the creation of the LISA hack space element on the program.

ND: We're trying a bunch of new things this year; we have new workshops for metrics and clouds. We have amazing content at both of these. We've added the new labs space, where people can do hands on activities. I'm also quite proud of the talks that we've recruited; we have good coverage of several new technologies that could really change systems and operations in the next few years. I think that it is important to experiment with new formats; it helps to keep the conference fresh.

BC: What was your first LISA and what brought you there?

ND: My first LISA was 2002 in Philadelphia. I actually showed up specifically for the configuration management workshop. We were building a variety of system management tools for some early beowulf clusters, including configuration management. It was my first exposure to the config community, and had a large impact on my career.

KS: I had to go back and check, and it turns out to be 1998 in Boston. Several of my colleagues at Chalmers University of Technology had attended previous years and it was generally considered the premier place to go for systems researchers and practitioners.

BC: What's been the most surprising to you during the planning and coordination for LISA '13? 

KS: It shouldn't be surprising, but adding a part time job, which chairing LISA is, to an already busy day-job, is quite taxing, and we've been lucky to have both patient co-workers and conference organizers at times when we've not been able to do good for either of them.

ND: Yeah, it was good that we had each other as backups when our regular lives got hectic. Also, filling 6 days with good content is really hard.


Fortunately, our co-chairs have succeeded in finding great content for LISA '13. Once again, attendees are forced to choose between sessions they want to see and sessions they really want to see. Registration is still open at or on-site.