It’s going to be about AJAX, JavaScript, OpenSocial and other related open and specific to Google web technologies. Among the speakers Guido van Rossum, author of Python.

More info on TechPresentations.org: Google I/O 2008

They have call for papers running till April 11 so if you want to give a talk there, you can send them 500-word abstract.

While you’re waiting for more info on this years conference, take a look at last year presentations on TechPresentations.org:

• Building a Scalable Resource Mgmt System for Grid Computing given by Khalid Ahmed
• Lessons In Building Scalable Systems given by Reza Behforooz
• Lustre File System given by Peter Braam
• SCTPs Reliability and Fault Tolerance given by Brad Penoff, Mike Tsai and Alan Wagner
• Scaling Google for Every User given by Marissa Mayer
• System Abstractions for Handling Large Datasets given by Jeff Dean
• Using MapReduce on Large Geographic Datasets given by Barry Brumitt
• VeriSign’s Global DNS Infrastucture given by Patrick Quaid and Scott Courtney
• YouTube Scalability given by Cuong Do

[via Google Research Blog]

• System Abstractions for Handling Large Datasets
• Building a Scalable Resource Mgmt System for Grid Computing
• YouTube Scalability
• Lessons In Building Scalable Systems
• Using MapReduce on Large Geographic Datasets
• SCTPs Reliability and Fault Tolerance
  
• VeriSign’s Global DNS Infrastucture
• Lustre File System

Plus Marissa Mayer’s presentation Scaling Google for Every User which is more about user experience measurement and so on – not so much about technical scalability.

BigTable is a distributed storage system for managing structured data that is designed to scale to a very large size.

Interesting quotes from presentation:

• Scale is too big for commercial databases, they can’t also run on a cheap clustered servers.
• Features:
  • Distributed multy-level map
  • Fault tolerant, persistant
  • Scalabale (thousands of servers, megabytes of in-memory data, petabyte of disk data, millions/sec of r/w, efficient scans)
  • Self-managing (servers can be added/removed dynamically, servers adjust to load imbalance)
• Largest bigtable cells (data collections) ~200TB on over thousands of servers
• Built upon:
  • GFS
  • Workqueue (scheduler)
  • Lock service
  • MapReduce
• miltidimentional – row (e.g. url), col (attribute) = cell, inside cell time-based values for the cell.
• related rows (tablets) are located on the same machines for better performance
• load balancing moves tablets around
• tablets are replicated across multiple machines
• requests like “get recent X values” are possible
• columns can be configured to retain only X most recent entries
• locality groups to partition tablets
• has huge logging problems
• a lot of opportunities for compression – time-shifted data is similar, many values are the same. Using BMDiff (dictionary-based compression) – encode ~100MB/s, decode ~1000MB/s; Zippy (LZW-like) – 179MB/s, 409MB/s
• Compression experiment results: web pages compress at 9.2%, links at 13.2%, anchors at 12.7%

Update: Luke Baker made screen shots from video with all slides (not really in the right order).

Interesrting presentation outlining major parts of infrastructure used by Google to run all it’s projects on.

These are main infrastructure system used at Google:

• GFS – Google file system (1+ PB, 1000+ machines, 1000+ clients, 10 Gb IO)
• MapReduce – distributed computation system using MapReduce programming model
• BigTable – distributed structured storage system

P.S. slides are not seen that good unfortunately so you’ll have to listen patiently.