Princeton S* Network Systems

I’m back in Princeton after spending the week in Indianapolis, Indiana for the first ACM Symposium on Cloud Computing (SOCC 2010). I’m posting here with a brief summary of each talk from Day 2 of the conference as well as some of my thoughts. These are my reactions to the presentations only, as I haven’t read most of the papers.

[See my previous post for Day 1]

Keynote 2

Building Facebook: Performance at Massive Scale
Jason Sobel (Facebook)

Jason gave a great presentation about the architecture at Facebook and some of the lessons their engineering team has learned in the last several … Continue Reading

I’m currently in Indianapolis, Indiana for the first ACM Symposium on Cloud Computing (SOCC 2010). I’m posting here with a brief summary of each talk at the conference as well as some of my thoughts. These are my reactions to the presentations only, as I haven’t read most of the papers.

[See my next post for Day 2

Keynote 1

[Note: My SIGMOD paper was being presented opposite the first keynote, so Steve Ko who is also at the conference wrote this first summary.]

Evolution and Future Directions of Large-Scale Storage and Computation Systems at Google
Jeffrey Dean (Google)

Jeff … Continue Reading

This year’s HotNets workshop was held over the past two days in the faculty club at NYU; it was nice being on old turf.   The HotNets workshop has authors write 6-page “position” or “work-in-progress” papers on current “hot topics in networking” (surprise!).  Tucked into a cosy downstairs room, the workshop was nicely intimate and it saw lots of interesting questions and discussion.

One topic that was of particular interest to me were new ideas about datacenter networking; HotNets included two papers in each of two different research areas.

The first thematic area was addressing the problem of bisectional … Continue Reading

In the previous post, I discussed how CoralCDN implemented bandwidth restrictions that were fair-shared between “customer” domains. There was another major twist to this problem, however, that I didn’t talk about: the challenge of performing such a technique on a virtualized and shared platform such as PlanetLab.  While my discussion is certainly PlanetLab-centric, its questions are also applicable to other P2P deployments where users run peers within resource containers, or to commercial hosting environments using billing models such as 95th percentile usage.

Interacting with hosting platforms

CoralCDN’s self-regulation works well in trusted environments, and this approach is used similarly in other peer-to-peer … Continue Reading