Fortunately, Microsoft blogger extraordinaire Dare Obasanjo made it to the conference and wrote up excellent and detailed notes on four of the talks:
- MapReduce, BigTable, and Other Distributed System Abstractions for Handling Large Datasets
- Using MapReduce on Large Geographic Datasets
- Lessons in Building Scalable Systems
- Scaling Google for Every User
Supposedly, "all sessions will be available on YouTube and Google Video after the conference." So far, I was only able to find one talk there, "Building a Scalable Resource Management System for Grid Computing".
The others may show up on Google Video in a few days. If they do, I will update this post with links to them.
If you are interested in this conference, you might also be interested in a few (    ) of my many older posts on large scale computing at Google that include links to other papers, presentations, and videos.
Update: Some other notes ( ) on the conference from Robin Harris at StorageMojo. [via Ewan Silver].
Update: Some more notes, this one from François Schiettecatte.
Update: Torsten Curst posted notes as well and was particularly impressed with the Amazon talk. As François Schiettecatte mentioned, a paper with at least some of the material from that talk, "Dynamo: Amazon's highly available key-value store", will be presented at the upcoming ACM Symposium on Operating System Principles.
Update: Dan Creswell has more details on the Amazon talk.
Update: Three more of the conference talks (  ) are now available on Google Video.
Update: Two more of the conference talks ( ) on Google Video, including the one by Marissa Mayer.
Update: In a more recent post, "Google Scalability Conference talks available", I listed all the talks that are available to make it easier to watch them.