7 takeaways from Velocity Europe

Taking a look at the current issues affecting the Web operations and performance space.

Editor’s note: The European edition of our Velocity conference wrapped up a few weeks ago, and now that the jet lag has passed I’ve had a chance to reflect on the talks and excellent hallway conversations I had throughout. And while I thoroughly enjoyed all the sessions I introduced, one of the downsides to being a chair is that I can’t attend all the other sessions at the same time. As such, I always look around for excellent dissections of the conference from other people; this summary by Peter Arijs from CoScale closely reflects some of the themes I saw, including a few of the standout talks.

velocity_barcelona_crop

November in Barcelona was full of action for web and big data practitioners, with the Velocity and Strata-Hadoop conferences and side events such as WebPerfDays and Papis.io. As a startup in the web application monitoring and analytics space, it was the perfect time to get a pulse on the state of the art, and talk to some of our clients and prospects. Below is a summary of personal take-away points from selected Velocity sessions and personal interactions.

1. Mobile and the need for speed

The continued increase in mobile traffic is certainly not surprising or new, whether it is via mobile web sites, hybrid apps or native apps. Over 20% of online shopping revenue comes in via mobile devices, and even more online shopping cycles start from mobile. Combined with the fact that sub-standard performance remains one of the top reasons for cart abandonment, this places increasing pressure on merchants and other online businesses to build fast mobile web applications. A large number of sessions at Velocity discussed techniques to optimize web apps for mobile, as well as monitoring techniques for mobile apps.

2. Content Delivery Networks: benefits & challenges

CDNs held a prominent spot at Velocity with most of the leading CDN vendors having a presentation and/or a booth at the exhibition. While CDNs are well known and suited for offloading static content from content-heavy web applications, some CDN vendors are also looking into new invalidation and purging techniques to serve more dynamic traffic from CDN nodes closer to the end-user, as Hooman Beheshti mentioned in his talk.

Based on our conversations at the conference, the ability to quantify the benefit of a CDN, from a performance, capacity and cost point of view, remains an important question. Understanding which part of your IT infrastructure is used by static and dynamic content, and how it affects performance, is something where a good monitoring and analytics tool can come in useful.

3. 3rd party content and how it affects your site’s performance

Besides the typical static and dynamic content on web pages, third-party content is also an important, but often overlooked component. Social widgets, advertising, and tracking and analytics contribute to the performance of web pages and, in the worst case, they represent a single point of failure. Deferring or asynchronous loading can optimize performance but, before that, site builders should reflect on what third-party components they really need, and research the 3rd party providers they work with. See @AndyDavies What are Third-party Components Doing to Your Site’s Performance? and @tameverts Everything You Wanted to Know About Web Performance (But Were Afraid to Ask) for more.

4. Monitoring tools need to adapt to changing infrastructure models

Besides CDNs, monitoring vendors took up another large chunk of the exhibit floor, with the focus clearly being on more self-serving monitoring tools, served as a SaaS solution. This includes synthetic testing and RUM solutions, as well as SaaS products for monitoring servers and applications inside the datacenter or cloud. Another new trend in the monitoring landscape is the increasing use of analytics in monitoring tools to automate problem detection and resolution through machine learning techniques, as also mentioned in a previous blog post, Analytics is the New Monitoring.

5. Anomaly detection: from reactive to proactive

Following up on the previous topic, anomaly detection is one of the techniques that can help in the early detection of performance problems without the need for the user to set up manual alerts. Anomaly detection is considered as a key enabler to detect problems that reside in the interactions between the many components that today’s distributed applications consist of. A presentation by @arun_kejariwal from Twitter provided some insight into the techniques they are using for anomaly detection, including time series decomposition to filter out trends and seasonality, and statistical techniques to identify outliers (as also explained in this blog on the topic). Other sessions touched on this topic too.

6. Contextual alerting to combat alert fatigue

Although anomaly detection can help in replacing manual alerts, people still want to be alerted about performance issues affecting their end-users. However, alert fatigue is a major problem, due to the overload of usually poorly documented alerts. To combat this, alerts should include as much context as possible, including severeness, history, business impact, related events, downstream effects, possible resolution, etc. During his presentation It’s 3AM, Do You Know Why You Got Paged?, Etsy’s @Ryan_Frantz summarized this as follows: “If a computer wakes me up in the middle of the night, it better do some work on my behalf first”. Well said!

7. Collaboration and empathy: key factors to address the human aspect of IT

Although Velocity scores pretty high on the ‘geek scale’, not all sessions are deeply technical. In his DevOps-inspired talk Collaboration Beyond the Tech Silo, @jystewart emphasizes the need for collaboration, communication and empathy across the entire organization. Sessions such as Life After Human Error by @StevenShorrock and Embracing Your Personal Apocalypse by @wbspjr gave a glimpse of the human aspect behind errors and recovery from those.

Wrap-up

Velocity remains the premier web operations and performance conference, with an excellent speaker line-up and a great place for sharing thoughts and ideas with peers. The above topics represent a personal selection of my Velocity experience. There were many more interesting sessions and topics, so feel free to add your impressions below.


This originally appeared on coscale.com as 7 Things to Remember about Velocity Europe, besides Sun and Tapas; it is republished here with permission.

tags: , , ,