Velocity Conference Review (London 2018)
Velocity is a technical conference focused on DevOps, web operations, performance, and the professionals that maintain and develop those services and applications. Velocity is organized by O’Reilly Media, who you may recognize from their books on engineering, videos, and other training materials. At Velocity, you’ll build skills to meet the pressing demands of your work, hear from your peers, meet with industry leaders, and get real knowledge that you can take back and apply right away to your work. Through expert-led presentations, you’ll learn to build high-performance, resilient, and secure systems. Whether you’re working on cloud migration, monitoring and observability, scalability, serverless, container orchestration, or leadership skills, you’ll get the most critical, up-to-the-minute knowledge from Velocity’s network of experts This year London was hosting the conference from October 31 to November 2, 2018.
The attendees at Velocity represented the diversity of the development and DevOps industry. I met everyone from software developers to system administrators to dedicated DevOps engineers. Like with most conferences of this size, Velocity offered a multi-track session schedule with several topics and themes including:
- Building Secure Systems
- DevOps and SRE
- Microservices and Containers
- Monitoring, Observability, and Performance
- Leadership and Career Growth
Google, Microsoft, GitHub, and other major players in the software industry were among the companies represented by the great line up O’Reilly brought together.
While there were many wonderful sessions at Velocity I definitely had a few favorites:
Chaos engineering bootcamp (Gremlin)
Speaker was talking about the concept of chaos engineering and how would that help companies to understand and avoid production issues by testing the resiliency of the system in advance. They made a correlation between Vaccine in human body and Chaos engineering in software. The same way that Vaccine is generated by an agent that stimulates the body’s immune system to recognize the agent as a threat, destroy it, and to further recognize and destroy any of the microorganisms associated with that agent that it may encounter in the future, a chaos in software systems will vaccinate the system for future failures.
Speaker also went through some tools and practices you need to implement chaos engineering in organizations. It helped to learn new ways to use chaos engineering within engineering organization and discover how other companies are using chaos engineering — and the positive results they have had using chaos to create reliable distributed systems.
Building evolutionary infrastructure (ThoughtWorks)
Speaker talked about how to make a robust infrastructure by using the right set of tools and best practices and avoid building up an infrastructure that is hard to maintain and change during time. He described the philosophy behind Infrastructure as code and emphasized the advantages of the approach. Speaker was the author of the O’Reilly book Infrastructure as Code
Topics include The role of different types of tools and technologies in an evolutionary infrastructure. Implementing automated infrastructure testing and validation. Designing pipelines to deliver changes across environments and into production. Handling data and configuration in a system where infrastructure is dynamically rebuilt, expanded, and contracted. Advanced patterns for structuring and integrating infrastructure projects and pipelines in larger systems used by multiple teams.
The keynotes happened in day 2. There were a lot of different talks from well-known speakers. Some of them that caught my attention the most were:
Kubernetes: Good or Evil? — The Ethics of Data Centers — Anne Currie (Container Solutions)
The speaker touched on how responsible we are like people who work in tech companies to care about the ethical aspect of our work. He touched on Stackoverflow’s research that was asking if a developer being asked to write an unethical code would you do that?! and %40 responded YES, and the other question was if you write an unethical code, would you personally feel responsible for that? and the answer was %80 said NO. She then started from that point to emphasize our responsibility toward our code and our systems. One of the options which were suggested was to run our code and services on sustainable servers provides by different cloud providers like Google, Amazon, Azure and going away from DCs which are consuming lots of energy and making the air pollution.
Incognito mentorship (Github)
The speaker talked about career progression and what would a career success mean! She talked about every day is a challenge, and how every challenge is novel. Learning dogs your every step. How could it not? When we know next-to-nothing, it’s either grow or fail. She took some nice example of success from people out of the tech industry and applied their methods of success to software companies. The comparisons and how she related everything at the end of the talk was so impressing and made me think a lot!
This was a nice talk regarding challenges people face on monitoring and logging on serverless architecture. He talked about how important is the execution time when we enter into the serverless world and how we should shave unnecessary processes when running tasks in serverless and at the same time having enough observability around what has been executed in serverless. he followed some good examples and best practices and made some good advices to people that are willing to move to serverless.
The Second rounds of keynotes happened in day 3. these were my favorite talks among them:
The Misinformation Age (Two Sigma Investments)
This talk was mostly how companies are turning to data-driven recruiting and hiring strategies and how would that affect peoples. The speaker talked about different stories that how an automated hiring process would fail to hire the right candidate and how companies can getter better on that. She touched on the fact that we can only learn from our false positives and we can never learn about our false negatives and that means we can only learn from candidates that been flagged as positives and end up not being a good candidate and we can’t learn from candidates that been flagged as negative and ended up going to other companies and being a successful person. At the end what she was suggesting was to always think about candidate first and put company second.
What changes when we go offline first? (University of Cambridge)
The talk was about the challenges we face when we design a distributed application and is suppose to work both offline and online and how synchronization will become a big deal! He explained why offline systems are important and described examples of where applications should work both in offline and online mode. At the end, he introduced an opensource library that addresses this issue on mobile applications and makes it easy for React developers to integrate it into their apps. More info on the library here
I really enjoyed the talk as it was reminding me what we have done in Skyscanner to migrate all our services to the public cloud and what challenges we face by that time and made me feel like we were not alone in that journey and other people experienced the same sort of challenges. Apart from that what made the talk more interesting was the lesson learned during the migration and what wrong assumptions being made before migrating to the public cloud. Topics included: Lift and shift versus refactoring and shift, Using the public cloud as IaaS versus PaaS, Securing operations in the public cloud, Solving for ever-evolving cloud services, Solving for high availability and disaster recovery (i.e., should you be in one AWS region or multiple ones?)
Final thoughts on Velocity Conference
Velocity definitely hit the mark when it comes to putting on a great technical conference. The diverse group of amazing speakers and engaging sponsors. Organizers did a great job of providing the best environment for professionals and encourage them more and more to meet and share ideas.
If you are involved with DevOps, web/application performance, or just want to see a conference done right I highly suggest making it to the next Velocity Conference. There is a wealth of information and networking available at these events that you shouldn’t miss!