Presentations: Operations

Listed below are the preliminary Operations sessions and workshops confirmed to date. Additional sessions will be added as they are confirmed. Schedule subject to change.

Web Performance

Simon Nicoud (Neustar), Mark Watson (Neustar), Ian White (Neustar)
Free, open-source web performance tools make it possible to build a functional web performance testing system in under a day. Two open-source browser automation frameworks (Selenium and Watir) will be demonstrated and we will show how they can be extended to gather performance metrics in order to aid Performance Engineering, Operations, Development, and QA teams.
Chris Blizzard (Mozilla Foundation)
.
Jason Weber (Microsoft)
Internet Explorer has fundamentally changed browser performance by taking full advantage of modern PC hardware. We’ll look at how the new JavaScript compiler, new GPU powered graphics system, new layout engine, and new platform capabilities have made websites faster today while enabling the next generation of HTML5 applications through hardware acceleration.
Nicole Sullivan (Stubbornella)
HTML5 and CSS3 offer a host of new features - some of which can improve performance, and some which can be disastrous to your site speed. Big sites, especially, can't afford to alienate users by being too far out on the bleeding edge. In this talk, you’ll learn which features to use now and which to avoid. Learn to boost your site’s performance while maintaining fallbacks for older browsers.
John Rauser (Pinterest)
How many machines will we need for Q4? How big should our next datacenter build be? What are the odds this project will be completed on time? What does a network packet loss rate of 1% mean for my application? Do you feel like you ought to have a working knowledge of statistics, but don’t know where to begin?
Artur Bergman (Wikia/Fastly)
Track page speed, ad inventory, browser width/height and errors? All that is possible by realizing that Google Analytics lets you feed in arbitrary data and then allowing you to slice it to get new insights.
Martin Hunt (Game Closure), Bobby Richter (Mozilla)
During development of a cutting-edge HTML5-based game library, we solved several performance problems with canvas APIs across browsers and operating systems. We present common problems and solutions that developers using canvas will face. Topics include image manipulation techniques, browser and OS differences, upcoming HTML5 APIs, and JS optimization techniques for high-performance drawing.
Greg Schechter (YouTube), Phil Harnish (YouTube)
HTML5; it’s new, it’s awesome, and it’s powerful, but can it take down the champ of video distribution, Flash. Which technology's got the ability to bring cat video to the next level. This talk will cover the many challenges of both HTML5 and YouTube adapting to change video distribution on the web.
David Mandelin (Mozilla Corporation)
This talk will help JavaScript developers make their programs fast by popping the hood and explaining in concrete, intuitive terms how JavaScript engines run code. Developers will also learn about gotchas that can make JavaScript run much slower. Finally, the talk will preview future performance work, offering ideas about what new things may soon be possible in JavaScript.

Presentation

Marcel Duran (Google Inc.)
YSlow is back! We will be happy to have a demo of new features in next-gen YSlow.
Daniel Hunt (Shazam)
The advantages and disadvantages of a single web application codebase will be explored by migrating several core modules of Yahoo! Mail to a NodeJS server environment.
Karl Dubost (Opera)
Presentation: external link
All you wanted to know about Opera browsers but you didn't even know you could ask.

Presentation

Tim Caswell (HP webOS)
Has anyone ever told you have to choose between high-level programming languages and performance when writing backends to modern web services? If so, then they clearly haven't used NodeJS. This new project combines age-old systems wisdom and non-blocking I/O with the hot V8 JavaScript engine for a platform that taking over the world of server-side scripting.
Paul Roy (Microsoft), Alex Polak (Microsoft)
In this session we will share MSN’s current directions related to performance metrics and measurement systems, as well as how we assess the business impact of performance changes. We also will share the results of several case studies, including the impact on business metrics. Lastly, we will discuss the techniques we’re using to drive the performance mission inside of MSN.
Steve Souders (Fastly), Patrick Meenan (Google), Alois Reitbauer (Compuware APM Division), Sergey Chernyshev (truTV (Turner)), Craig Conboy (Blaze)
Go beyond the numbers and learn how to do more advanced analysis and testing with WebPagetest. Evaluate the impact of third-party content, script multi-step paths and see the actual user experience all without changing site code.
Billy Hoffman (Zoompf)
Presentation: external link
While the web is full of images, we restrict ourselves to <15% savings using lossless optimization tools. With lossy optimizations we can achieve 30-70% savings. “Losing data” sounds scary. See how to intelligently approach images and their content with detailed guidelines for applying lossy compression to achieve substantially smaller file sizes while maintaining image quality and user experience
Julia Lee (Yahoo!, Inc.), Stevan Vlaovic, PhD (Yahoo! Inc.)
This talk gives an overview on how the advertising ecosystem, application architecture, ad integration, and global serving footprint holistically affect end to end performance and user experience. Ad investigation, instrumentation, measurement, and monitoring methodologies will be shared. The audience will also hear about interesting techniques to improve perceived performance and user experience.
Bryan McQuade (Google), Joshua Marantz (Google)
The Page Speed tools help web developers and site admins make web pages faster. We’ll cover new Page Speed features, such as optimizations for mobile devices and modern desktop browsers. We'll share learnings from mod_pagespeed, an Apache implementation of Page Speed Automatic, which rewrites sites using Page Speed rules. We'll close by exploring: Which rules have the largest impact on the web?
Patrick Meenan (Google)
Presentation: WebPagetest Update Presentation [PPT]
There have been a lot of user-facing features that have been added to WebPagetest over the last year and this talk will help you make sure you get the most benefit from the tool. The presentation will also cover information on running your own private instances of WebPagetest and use of the API for automation or integration with existing systems.

Operations

Mohit Lad (Thousand Eyes Inc)
How many locations should one serve content from to make their application truly fast globally? At what point does the improvement become marginal? These are questions that we answer in this talk using actual network performance data and arm the audience with the knowledge necessary to make informed decisions on network planning for optimizing application performance.
Theo Schlossnagle (OmniTI/Circonus)
Presentation: external link
We all know that web performance is key. It drives adoption, loyalty and revenues. While it is so important, real-time web performance metrics are often absent from operations culture. Which metrics are important to operations? How can I use these metrics to improve performance, availability and stability? Learn how to make operational sense out of this critical performance indicators.

Presentation

Jeff McCune (Puppet Labs)
Traditional methods of monitoring service performance have shortcomings due to the on-demand nature of cloud computing. Learn how to deploy MCollective to provide a self-organizing service monitoring and remediation.
Philip (Flip) Kromer (CSC), Dennis Yang (Infochimps)
Chef isn't just 'shell scripts but better', and Flume isn't just 'manages logs'. These are first-class superpowers for radically decoupling systems, and the more you decouple, the more benefits beyond the system diagram emerge. I'll show how these tools let you rethink team structure, hand interns the keys to the Ferrari, ignore every best practice in sight, and scale on a shoestring.
Mark Imbriaco (DigitalOcean)
The Heroku platform grew from hosting 0 applications to more than 120,000 without ever buying a server. We'll talk about the overall architecture of the platform and dive into some of the operational consequences of those architectural choices.
Gene Kim ("The Phoenix Project: A Novel About IT, DevOps, and Helping Your Business Win" )
I'm going to share my top lessons of how great IT organizations simultaneously deliver stellar service levels and fast flow of new features into production. It requires creating a "super-tribe", where development, test, IT operations and information security genuinely work together to solve business objectives as opposed to throwing each under the bus.

Presentation

Justin Sheehy (Basho Technologies)
We will face the reality that anything we depend on can fail, and in the worst possible combination. We will learn how to build systems that let us happily sleep at night despite this troubling realization.
Joshua Timberman (Opscode, Inc.), Adam Jacob (Chef), Christopher Brown (Opscode), Aaron Peterson (Opscode), Matt Ray (Opscode), Seth Chisamore (Opscode, Inc.)
System administrators love Chef because it gives them flexibility to integrate all aspects of their infrastructure such as monitoring and trending tools with applications. Software developers love Chef because it helps them take care of the muck so they can focus on writing great applications.

Presentation

Bryan Cantrill (Joyent, Inc.)
We describe our experiences developing DTrace-based system visualization ofNode.js in an early production environment: the 2010 Node Knockout programmingcontest. We describe the challenges of instrumenting a distributed, dynamic, highly virtualized system -- and what our experiences taught us about the problem, the technologies used to tackle it, and promising approaches.
Robert Treat (OmniTI)
It's great to talk about bringing dev tools to ops teams, or making developer's more aware of production issues, but your database is where the rubber meets the road. Join me in this talk where I'll outline common strategies used in many of todays leading devops shops to keep things running. It's not just about dba's anymore.
Philip Zeyliger (Cloudera)
What new challenges does managing and operating distributed systems pose? How is managing distributed systems fundamentally different than managing single-node machines? In this talk, we discuss the problems and work through the architecture of a system that eases lifecycle and configuration management of HDFS, MapReduce, HBase, and a few others.
Adrian Cockcroft (Battery)
Presentation: external link
Details of why Netflix is running on AWS, what is running where and how it was done. Lot's of practical details on re-architecting large scale web services to run in the cloud, and new operational patterns.

Presentation

Ron Pedde (Rackspace Hosting), Todd Willey (OpenStack), Matt Ray (Opscode)
OpenStack is a collection of open source technologies delivering a massively scalable cloud operating system.
Moderated by:
Mandi Walls (Opscode)
Panelists:
Mark Imbriaco (DigitalOcean), Matt Hackett (Tumblr), Teresa Dietrich (WebMD), Jake Loomis (Yahoo!)
A panel discussion featuring folks from sites which have had large outages in the past months. Discussion topics would cover causes, resolutions, process changes after the outage, cultural changes after the outage, and lingering after affects with regards to how the userbase reacted.
Jeremy Edberg (Netflix)
reddit.com does 1.3 billion pageviews a month, and that number grows by about 10% each month. Through that growth they've made some mistakes along the way. Some they have fixed, and some they are still paying the price. Come hear Jeremy Edberg, reddit's head of technology, speak about their successes and failures.
John Adams (Twitter)
Interconnecting many systems to express a single function is often a difficult task. What seems like a simple problem, becomes a difficult one when trying to serve millions of users at the same time without failure or downtime. Timeouts, redundancy, humans, and high levels of concurrency are a challenge, and I'll walk through problems and solutions encountered while scaling twitter.com.
Kate Matsudaira (SEOmoz)
What do you do when everything falls apart during your product launch? This talk will cover tools for surviving the worst, and making the most of it. We'll look at post-launch checklists to help recover rapidly, managing expectations to instill confidence in your customers and management, keeping up team morale, and successful post-mortems.
Sarah Novotny (NGINX)
Modern web stacks have many different layers that can benefit from caching. From raid card caches, disk caches and CPU caches to database caching, application code caching and rendered page caching there is a lot to consider when optimizing and troubleshooting. Let's investigae what's redundant, what's optimal and where your applications might benefit from a closer look.
Jake Loomis (Yahoo!)
This year the Yahoo! frontpage went dark. Tweets rang out, "What happened? Yahoo! is never down. I can't remember the last time this happened." What enabled www.yahoo.com's remarkable record of reliability? This session will cover the top 5 tricks that contributed to the stability of Yahoo!'s frontpage and a description of what caused the eventual downtime.
Jason Cook (Wikia)
In the pursuit of making our site fast, efficient, and resilient in the face of failure we have chosen to use geographically dispersed data centers. In this session I will cover the tools and techniques we used to extend MediaWiki to render pages out of multiple geographically dispersed data centers.

Velocity Culture

John Allspaw (Etsy)
Presentation: external link
Getting tight operationally means strengthening the resiliency of both your stack _and_ your organization's response to issues that arise. Outage postmortem meetings done wrong can be stress-filled blamefests, and done right can be collaborative illustrations of Resilience Engineering. I'll use Etsy.com examples to illustrate sticky topics such as Root Cause Analysis and Human Error.

Presentation

Damon Edwards (DTO Solutions), Lee Thompson (Lee Thompson), Alex Honor (Simplify Ops), John Willis (Enstratius)
Presentation: external link
The old model of "productionizing" a release after it leaves development is flawed and often leads to disaster. Using lessons learned inside leading web companies, Lee and Damon make the case for bringing fully automated deployment and testing, along with a "production mindset",to every step of the application lifecycle, from early development environments through to customer-facing environments.

Presentation

Adam Jacob (Chef), Jesse Robbins (OnBeep, Inc.)
Presentation: external link
Adam Jacob returns with Choose Your Own Adventure 2: Electric Boogaloo ;-)
Joshua Timberman (Opscode, Inc.), Adam Jacob (Chef), Christopher Brown (Opscode), Aaron Peterson (Opscode), Matt Ray (Opscode), Seth Chisamore (Opscode, Inc.)
System administrators love Chef because it gives them flexibility to integrate all aspects of their infrastructure such as monitoring and trending tools with applications. Software developers love Chef because it helps them take care of the muck so they can focus on writing great applications.

Presentation

Patrick Debois (Jedi BVBA), Israel Gat (Cutter Consortium), Andrew Clay Shafer (Pivotal)
Ah, great technical sessions. Ready to get back to work to try that uber geeky tool? Consider the folllowing questions: How do you keep your organization scalable, adaptable, secure and performing? How do you check the health at the human level? Or ‘measure’ the devops gap to make sure the collaboration stays at a top-notch level. Ideas and insights from recent client engagements.
Sixty-five engineers at Etsy deploy code more than 30 times a day with continuous deployment. Learn about the tools we've put in place to monitor our infrastructure and application health throughout the day, and how this is a shared effort between our engineering and operations team effort.
Moderated by:
Mandi Walls (Opscode)
Panelists:
Mark Imbriaco (DigitalOcean), Matt Hackett (Tumblr), Teresa Dietrich (WebMD), Jake Loomis (Yahoo!)
A panel discussion featuring folks from sites which have had large outages in the past months. Discussion topics would cover causes, resolutions, process changes after the outage, cultural changes after the outage, and lingering after affects with regards to how the userbase reacted.
Kate Matsudaira (SEOmoz)
What do you do when everything falls apart during your product launch? This talk will cover tools for surviving the worst, and making the most of it. We'll look at post-launch checklists to help recover rapidly, managing expectations to instill confidence in your customers and management, keeping up team morale, and successful post-mortems.

Mobile Performance

Josh Fraser (Torbit)
In this talk, Josh Fraser (CEO of web performance company, Torbit) will discuss some of the tricks for optimizing a mobile website for performance. As we all know, performance on mobile is painfully slow. This talk will look at the differences between devices and how to do mobile testing. We will discuss resizing images, lazy loading, preloading, and knowing when to inline or outline css.
Ariya Hidayat (Sencha)
While many developers use proven best practices to ensure that the performance of web applications is acceptable, further tuning and tweaking are useless without real measurement data. A wide range of analysis tools are available for major desktop browsers, however the situation is different on the mobile space as there is very little instrumentation possibilities.
Mike Weider (Blaze)
In this presentation we’ll share our research comparing mobile browsing of the top 500 websites from the three leading smartphone platforms. We’ll show which was faster when, compare devices and networks, and give some advice on building fast mobile web applications. We’ll also share a free mobile browsing assessment tool to use on your own site.
Maximiliano Firtman (ITMaster Professional Training)
Mobile Web is here. Smartphones and Tablets are here. Does typical WPO techniques still apply on mobile browsers? How can we measure performance on mobile browsers? How accurate are the solutions available? In this session, we will discuss one of the most important topics in WPO during 2011: Mobile Web.
Dylan Schiemann (SitePen, Inc.), Kris Zyp (SitePen, Inc.)
Learn about the exciting new WebSockets API and how it can make your real-time web application really fast. We'll explore solutions built on top of Node.js with a focus on how to remove barriers of complexity and scalability, and a variety of Comet-related topics.
Rajiv Vijayakumar (Qualcomm Technologies, Inc.)
An in-depth look at two critical factors affecting mobile Web browser performance: first how to optimize browser performance on 3G/4G mobile networks, including managing higher latencies and the cost of bringing radio connections up and down; second looking at mobile browser software architecture for Android, including reduced cache sizes, and support for parallel connections and pipelining.
Yehuda Katz (Tilde Inc)
It's a myth that the web stack is incapable of performing well, but writing fast applications with a lot of logic on the client-side can be a lot like walking through a minefield. For the past five years, SproutCore has been carefully identifying and defusing these land-mines for high-profile, large-scale applications. Learn about some of these techniques, and how you might apply them to you apps.

Products & Services

Stephen Ludin (Akamai)
Feeding a need for ‘as-it-happens’ interaction, many popular Web sites, Web- based applications, and social networking platforms incorporate real-time updates into their user interfaces. However, providing these real-time updates to a large and distributed user population through a server-sent events mechanism can create significant scalability challenges. *This session is sponsored by Akamai*
This session is sponsored by Apica
This session is sponsored by Cotendo
Lew Cirne (New Relic)
New Relic persists over 90,000 metrics per second while delivering an average page load time of 1.5 seconds. In this session I will discuss how good architecture and good tools can help you handle an extremely large amount of data while still providing extremely fast service. I'll show you how New Relic scales to support customer growth, how we monitor our system, and what traps to look out for.
Ido Safruti (Akamai), Mike Belshe (Google)
SPDY was proposed by Google back in November 2009 to reduce the latency and load time of web pages. Cotendo implemented the server side, and extended our proxies to support SPDY, providing SPDY to HTTP “translation”. It really speeds things up. Hear insights from our implementation, optimization of SSL-based traffic and present performance data both from Google’s and our customers’ deployment.
Sven Hammar (Apica Systems)
Sven will be speaking about how to not become another bad web story. We’ve all heard the stories of sites crashing and performing poorly; it's only a matter of time before another one hits the headlines. Sven will demonstrate how to work with selenium scripting, utilizing the latest tools and expertise that result in “Rapid Actionable Insights.”

Ignite

Seemingly out of no where the traffic comes. First SYN packets, then ICMP pings, random recursors sending rogue DNS queries, and then millions of slow HTTP connections that grind your web app into the ground. No performance optimization matters if you can't reach your router, let alone your site. Come listen to the story of how bad it can get when it gets really bad.
How do you make sure you're hiring engineers with the talent and skills you need? Interviewers often go into interviews woefully unprepared. The result is candidates who are bored by answering the same questions for every interviewer, and a set of interviewers who still aren't sure about their candidates at the end of the day.
Understanding the basics of queuing theory is fundamental in understanding websites' capacity and performance. And it turns out, Little's Law can be intuitively explained using real-life metaphors from...Disneyland. This 5-minute talk will teach you all about this important theoretical concept with very practical applications.
Come see all the hijinks as Botchagalupe plays the role of Sally ("The Change Manager"). Watch Sally dance around all the ridiculous tricks applied to circumvent her large insurance companies change policy. Be on the look out for some great cameos staring as Bob ("The Clueless Operations Manager") and Betsy ("The Oracle").
Unemployment is still at a nationwide low, but yet it's so hard to find a home to rent in the city of San Francisco. Here are some stories of how a group of us, who independently moved to San Francisco, on how we found our respective places to live and learned to play the rental game. These will be great tips for anyone looking to relocate to the City by the Bay.
Everyone talks about agile development and lean startup methods bringing you things "just in time". But all too frequently, things aren't just in time, especially in a more operational context. Instead, they happen after the fact and after you really would have wanted to have had them.
Travel photo slide show with a twist. Between March of 2009 and September of 2010 I visited Holland, Turkey, England, Mexico, Brazil, India, China, and Germany. Five minutes should be enough to hit the highlights of what we saw, the awesome drinks we had, and not long enough to bore the pants off the audience.
A brief story of what magic was present at Lucasfilm in the creation of Pixar, what the boys who started it learned from each other, and a hypothesis of why they are unnaturally excellent. Lessons for any tech startup and it's just plain cool to see. (The story and rare images come from my book DROIDMAKER.)
  • Keynote Systems
  • Cisco
  • Google
  • Neustar
  • Betfair
  • Cotendo
  • Rackspace Hosting
  • Akamai
  • Apica
  • dynaTrace
  • Equinix
  • Facebook
  • New Relic
  • Opscode
  • Salesforce.com
  • Yahoo! Inc.
  • AppDynamics
  • Aptimize
  • Blaze
  • CDNetworks
  • Cedexis
  • Citrix Systems
  • Compuware Corporation
  • Dyn Inc.
  • F5 Networks
  • Heroku
  • Percona
  • Quest Software
  • Schooner Information Technology
  • SiteSpect
  • Splunk
  • Strangeloop
  • WatchMouse
  • Zeus Technology
  • Neustar

Sponsorship Opportunities

For information on exhibition and sponsorship opportunities at the conference, contact Yvonne Romaine at yromaine@oreilly.com

Download the Velocity Sponsor/Exhibitor Prospectus

Contact Us

View a complete list of Velocity contacts