Presentation Files

Presentation files will be made available after the session has concluded and the speaker has given us the files. Check back if you don't see the file you're looking for—it might be available later! (However, please note some speakers choose not to share their presentations.)

This talk focuses on tips for making application logs more useful to operations staff, particularly after a pager goes off at 3 a.m. We'll look at methods for managing logs, separating problems from metrics, and working with developers with a goal toward faster recovery after problems and automation of simple fixes.
Presentation: AOL PageTest Presentation [PPT]
More information coming soon
For most operators, BGP is black magic. So is scaling past a single datacenter. Not anymore. Using OpenDNS' global network as a real example you will get a complete picture of what's required to build and run a global fault-tolerant network spanning multiple datacenters. This includes how to load balance using standard routing protocols like OSPF and BGP (Anycast).
By utilizing a demonstration company running in EC2, we'll show you what all the components of an automated infrastructure are, how they work together to solve real problems, and why they provide an important competitive advantage.
Deciding which web technology is right for you and your new web application? You don't want to miss this session. We'll provide our experiences creating and testing a social networking application using three technologies: PHP, Ruby on Rails, and Java. All implementations are tested with components commonly used in large scale web sites, i.e., MySQL and Memcached.
Presentation: Capacity Management Presentation [PPT]
Planning and managing capacity for a fast-growing website can be a balancing act between buying too little/late, and too much/soon. Your process of capacity planning should be *adaptive*, *adjustable*, and include more than just system statistics. Measurement, architecture, and economics are all equally important to having your site perform.
Modern Web 2.0 applications, such as GMail, Live Maps, Facebook, and many others, use Ajax to push page generation and content manipulation to the client web browser. This improves the responsiveness of these network-bound applications. We present Doloto, a system that analyzes application workloads and automatically performs code splitting of existing large Web 2.0 applications.
In this talk, we will present Eucalyptus - an open source implementation of cloud computing that is interface compatible with's EC2.
Souders' book, "High Performance Web Sites," describes the 14 best practices he developed while working as the Chief Performance Yahoo!. YSlow, the Firebug extension he created, codified those best practices. Now working at Google, Souders discusses the next set of best practices he's discovered, including the best ways to load JavaScript and where to place (and where not to place) inline scripts.
Presentation: Fiddler Presentation [PPT]
The Fiddler Web Debugger enables capture, measurement, replay, and modification of HTTP and HTTPS traffic from virtually any application, including all major browsers. A highly-customizable freeware platform, Fiddler is extensible using either Javascript or any .NET language.
More information coming soon
Taking off from where the Yahoo! Exceptional Performance rules leave off, find out what it takes to design, implement, and test high performance Ajax applications, including common (and less common) techniques and patterns, and a case study.
How does Hotmail balance between the demands of a rich UI that is also performant? This presentation covers the lessons learnt, how to measure performance, and the trade-offs between Web 1.0 and Web 2.0.
Presentation: IE 8 What is Coming Presentation [PPT]
More information coming soon
Do you want to speed up your web pages and lower you hosting bills? Do you want to do this with little or no code changes while keeping the same beautiful UI? Come and learn about 7 simple steps to put your web images on a diet. Take off (and keep off) all the bytes you put on under the stress of chasing that next killer feature.
At Netflix we have implemented round trip tracing to get an accurate measure of the end user performance experience. With this data in hand we are marching through the set of YSlow recommendations + others and tracking the measured improvements.
Have you ever wondered how fire departments organize themselves on the fly to deal with a major incident? How they effectively coordinate the efforts of multiple agencies? How they evolve the response as the incident changes? Brent Chapman will introduce the Incident Command System (ICS), and discuss how it can be applied to IT events, such as security incidents and service outages.
More information coming soon
Jiffy is a new open-source end-to-end system built by the team for instrumenting your web pages, capturing client-side timings for any event that you determine, and storing and reporting on those timings. Learn about the motivations, goals, lessons learned, and how to participate.
Get an early look at KITE - the Keynote Internet Testing Environment. KITE tests and analyzes the performance of web applications from your desktop and locations around the world. KITE allows advanced scripting, detailed network performance and DOM analysis, and the ability to gather data from hundreds of geographic locations worldwide.
More information coming soon
The LinkedIn Communication Platform is built 100% on top of Java technologies and is responsible for handling all user communications and the distribution of network events. Learn about the architecture, implementation, and scalability challenges in building the communication platform for the world's largest and fastest growing professional networking service.
Fast web pages are critical to a good user experience, but what is "response time" and how should it be measured? This panel discusses what parts of page loading should be measured and the importance of perceived versus instrumented speed.
The Mobile Web is the wild west when compared to desktop browsers. There is little information about how mobile browsers handle the factors that impact performance. Learn about the tests that Cloud Four has conducted on mobile browsers and our conclusions about how to build fast-loading mobile web pages.
Presentation: Performance Metrics Presentation [PPT]
Sharing performance data across an organization is challenging. People want one number or one color (green or red), but often the key issues are only revealed when you look deeper. The panel of experts reviews examples of correlating performance metrics with business metrics, critical to anyone evangelizing better performance within their company.
Presentation: Performance Plumbing Presentation [PPT]
As a web site scales, the plumbing (i.e., network) creates opportunities for performance gains. This session looks at evolving a web site's plumbing from a single location to a multinational network with anycast and caching services.
Firebug is well known as a powerful, easy-to-use web-site inspector. Less well known are its performance analysis abilities. In addition to straightforward JavaScript code on web pages, some sites use dynamically loaded JavaScript or use JavaScript to interpret AJAX JSON strings.
More information coming soon
As your web application grows, no matter how you optimize it, you will come to the point when a single MySQL server is unable to keep up. Sharding and Replication are powerful techniques that allow you to scale your application to use Multiple MySQL servers.
Shared Dictionary Compression over HTTP protocol (SDCH) aims at reducing data redundancy across HTTP responses. The protocol is meant to work with current schemes (gzip, deflate) to further compress the HTTP responses. This protocol is different from original proposed rfc3229 (differential compression), as it does not require the browser to cache the last version of pages.
Presentation: SOX Compliance Presentation [PPT]
If your company is (or hopes to be) publicly traded, you're going to have to deal with Sarbanes-Oxley compliance. This brief presentation will teach you how to use compliance requirements and process to improve existing operational issues.
Presentations: Wikipedia Presentation [ZIP],
Wikipedia Presentation 1 [ZIP]
More information coming soon
You’ve seen how Google and Yahoo keep running 24/7, now it’s time to learn how Microsoft pulls it off…running Windows. This session will spill the beans on Microsoft’s Autopilot datacenter automation system and let you in on some of our future plans for running our ever-expanding server clusters
  • Akamai
  • Google
  • Hyperic
  • Sun Microsystems
  • 3Tera, Inc
  • Coradiant
  • Keynote Systems
  • Limelight Networks, Inc.
  • OpSource
  • TechRepublic

Sponsorship Opportunities

For information on exhibition and sponsorship opportunities at the conference, contact Sharon Cordesse at

Press and Media

For media-related inquiries, contact Maureen Jennings at

Contact Us

View a complete list of Velocity contacts.