Build resilient systems at scale
October 12–14, 2015 • New York, NY

performance conference sessions

9:00am–12:30pm Monday, 10/12/2015
Paul Calvano (Akamai Technologies), Rakesh Chaudhary (Akamai)
Slides:   external link,   external link
A rudimentary understanding of caching will ensure a decently performing web application. With the advanced topics covered in this tutorial, you can make that web app FLY! The caching ecosystem includes browser cache, html 5 app cache, reverse proxy cache, and CDN - a caching strategy that maximizes the potential of this ecosystem can push your web app performance from decent to best-in-class.
11:20am–12:00pm Tuesday, 10/13/2015
Tim Kadlec (Snyk)
Slides:   external link
Proxy browsers get a bad rep. "They’re antiques." "They’re crappy browsers." But a deeper understanding of what they accomplish, and what they do behind the scenes, reveals that they’re actually ingenious and capable of providing a significant boost in performance. Far from becoming outdated, they’re uniquely positioned to bring the web to the next wave of people and devices around the world.
3:40pm–4:20pm Tuesday, 10/13/2015
Charlie Park (Slack)
Newer CSS layout techniques like Flexbox and Grid Layouts can provide surprising benefits to both engineers and end users. We'll demystify these approaches, talk about when to use them (and when not to), and go over real-world examples of their use at IFTTT.com (and how they speed up rendering times).
3:40pm–4:20pm Wednesday, 10/14/2015
Nathan Bower (Zillow)
Slides:   1-PPTX 
Performance budgets lend a relatively simple safety net that enables designers and developers to build engaging features, while supporting optimal performance or reaching for higher performance goals. I'll discuss how to create good performance budgets, why it's important to have them in place, and tell a true story of how Zillow was able to respond quickly in a case when budgets were exceeded.
11:00am–12:30pm Monday, 10/12/2015
Philip Tellis (SOASTA), Nic Jansma (Akamai)
Slides:   external link
Investigating performance problems often requires more than one tool to nail down the problem. In fact, using the wrong tools can often point you completely in the wrong direction. In this talk, we will walk through a real situation we recently dealt with to identify the cause of a performance problem using RUM, Synthetic, and packet captures.
1:30pm–3:00pm Monday, 10/12/2015
Radu Gheorghe (Sematext Group), Rafał Kuć (Sematext Group)
Slides:   external link
The ELK stack (Elasticsearch, Logstash, Kibana) is all the rage now when it comes to centralizing logs. This talk will cover all the basics, and will quickly change to production-tips mode: we'll show you how to tune and scale both Elasticsearch and its indexing pipeline for handling huge loads with minimal overhead.
1:15pm–1:55pm Tuesday, 10/13/2015
Ragnar Lonn (Load Impact), Daniel Stenberg (Mozilla)
Slides:   external link,   external link
The new HTTP/2 protocol is mostly focused on web performance. So what kind of performance improvement can you expect by switching to HTTP/2? Daniel Stenberg (one of the IETF's architects of HTTP/2) and Ragnar Lönn (founder of Load Impact) explain the key improvements within HTTP/2, and unveil a new tool developers can use to see how their applications will function in the new protocol.
4:35pm–5:15pm Wednesday, 10/14/2015
Philip Tellis (SOASTA), Nic Jansma (Akamai)
Slides:   1-PDF 
It’s hard to tell when a new component of an SPA was requested as a result of an intentional user action or something else. Add on the various ways of building SPAs, ranging from raw XHR to Angular, Backbone, and sometimes all of the above, and you have an interesting problem in traffic analysis. Find out how boomerang measures the performance of SPAs in a way that works for real websites.
1:30pm–3:00pm Monday, 10/12/2015
Tammy Everts (SpeedCurve), Cliff Crocker (Akamai)
Slides:   external link
There’s no one-size-fits-all approach to metrics. In this session, we’ll walk through various metrics that answer performance questions from multiple perspectives — from designer and DevOps to CRO and CEO. You’ll walk away with a better understanding of your options, as well as a clear understanding of how to choose the right metric for the right audience.
2:10pm–2:50pm Tuesday, 10/13/2015
Baron Schwartz (VividCortex)
Slides:   1-PDF 
If you're monitoring a lot of data in an event stream (as VividCortex does query traffic monitoring, for example), sampling the stream usefully can be a hard problem. There are all sorts of edge cases and bad consequences. This talk explains how a sketch - a probabilistic data structure - turned out to be the answer for us.
3:30pm–5:00pm Monday, 10/12/2015
Chris Love (Love 2 Dev)
Slides:   1-PPTX 
Web performance optimization techniques and practices are well defined, but rarely utilized to create fast, modern, single page web applications. In this tutorial Chris Love will demonstrate how to apply many common web performance optimization techniques and tricks to build a fast, native-like application user experience that end users desire.
4:35pm–5:15pm Tuesday, 10/13/2015
Yoav Weiss (Akamai)
Slides:   external link
Getting faster load times often means downloading resources as soon as possible, but sometimes the browser doesn't know early on what resources would be required for the page. This is where new standards, Resource Hints and Preload, come in. They give us the ability to clue the browser in on what's going to happen in the current page, and the next one. Learn all about them!
4:35pm–5:15pm Tuesday, 10/13/2015
The service oriented architecture (SOA) at Twitter comprises of hundreds of services. Each service pushes out code at velocity. In light of this, we developed a statistically robust approach to detect both functional and performance regressions. This is critical in order to minimize impact on the end-user experience. In this talk we shall walk through the challenges and how we addressed them.
3:40pm–4:20pm Wednesday, 10/14/2015
Sonia Burney (Akamai), Sabrina Burney (Akamai)
Slides:   1-PDF 
The challenge of maintaining a secure site is typically prioritized over creating a fast and optimal site, because of the focus on blocking malicious traffic at the origin. It is often forgotten that security is not only a concern at the origin, but also at the browser. And with that, there are new opportunities to safely secure the user experience while also optimizing the front-end experience.
1:30pm–5:00pm Monday, 10/12/2015
Björn Rabenstein (SoundCloud), Julius Volz (SoundCloud)
Slides:   1-PDF 
Prometheus is a popular open-source monitoring system and time series database written in Go. It features a multi-dimensional data model, a flexible query language, and integrates aspects all the way from client-side instrumentation to alerting. Participants in this tutorial will instrument a toy service and set up all critical components of the Prometheus ecosystem.
2:10pm–2:50pm Wednesday, 10/14/2015
Kent Alstad (Radware Canada)
Slides:   external link
We wanted to know whether a faster page load is always better, or whether a page that draws with higher processing fluency will result in increased user engagement. This session presents the results of our study, comparing how user engagement is affected by various performance improvements. See how our experiments on how humans digest web pages can improve your website’s user engagement.
11:20am–12:00pm Wednesday, 10/14/2015
Ellen van Keulen (Akamai)
Slides:   1-FILE 
Delivering a consistent experience on the web is challenging at the best of times. Delivering into China is even more difficult! In this presentation we’ll cover the challenges, the common pitfalls, the processes involved in delivering within China, and some practical steps to be more successful.
2:10pm–2:50pm Wednesday, 10/14/2015
Kyle Parrish (Fidelity Investments), David Halsey (Fidelity Investments)
Slides:   1-PDF    2-PDF 
How do you stress test a brokerage system in production if you can't risk orders processing, trades executing, or violating regulatory obligations? The answer used to be: "We don't." But the flash crash and other market anomalies exposed the risks inherent in not testing production. Hear what we learned as we built a way to do what had been written off as "too big to test."
4:35pm–5:15pm Tuesday, 10/13/2015
Tim Kadlec (Snyk), Ilya Grigorik (Google), Patrick Meenan (Google)
We're trying something new this year in New York. We've had a lot of interesting announcements recently, and we're picking one of them for a (friendly) debate. Google just announced AMP, or Accelerated Mobile Pages. Some say it's a performance win for everyone, others feel it's a slightly different take on the "walled garden" that is counter to a vision of an open web. Let the debate begin!

Stay Connected

Follow Velocity on Twitter Facebook Group Google+ LinkedIn Group

Videos

More Videos »

O’Reilly Media

Tech insight, analysis, and research