Velocity San Jose June 2017 was a gathering of the world's top CTOs, DevOps practitioners, SREs, systems architects, Cloud team leaders, and application developers. Their goal: define and share the best ideas and technologies for building a stronger and faster web. This video compilation provides a complete recording of over 90 inspirational keynote addresses, in-depth tutorials, and individual sessions delivered at this thought provoking event—more than 70 hours of material to view at your own pace.
Keynote highlights include Martin Woodward's (DevOps program manager, Microsoft) tale of how Microsoft transitioned 65,000 Microsofties to DevOps in the public cloud; Dianne Marsh's (director of engineering, Cloud Tools, Netflix) reflections on how things have changed (and stayed the same) in her 30-year software career; Kelsey Hightower's (staff developer advocate, Google Cloud Platform) recount of his experiences building cloud-native applications with Kubernetes and Istio; and Cliff Crocker's (VP Product, Akamai) talk on "The False Dichotomy of Finders Versus Fixers" (you have to see/hear it to understand it).
One of the best values in this compilation is unfettered access to each of Velocity's 10 in-depth tutorials (3.5 hours each). A few samples include training on how to build a multinode, secure Docker Swarm cluster; how to deploy and manage a multicontainer, highly available microservice app inside a cluster; a hands-on data science crash course for modeling and predicting the behavior of (large) distributed systems; and a course in Linux performance monitoring using BPF.
The real gold of Velocity has always been found in the precise and always on-point 40 minute sessions, and Velocity San Jose June 2017 was no exception. Here is a brief rundown of the amazing variety of topics and talks delivered in this compilation:
- Monitoring, Tracing, & Metrics (4+ sessions): includes Megan Antcil (Slack) on wrangling the best out of open source monitoring tools; Suman Karumuri (Pinterest) on PinTrace and the excitement of distributed tracing technology; and Christine Yen (Honeycomb) on why standard monitoring set-ups using pre-aggregated metrics aren't sufficient for complex systems.
- DevOps & Tools (6+ sessions): includes Laura Frank (Codeship) on the challenges (achieving quorum, leader elections, scheduling, etc.) of running a distributed system, and Sangeeta Narayanan on the lessons Netflix learned from operating a serverless-like platform at scale.
- Networking, Traffic, & Edge Management (6+ sessions): includes Matt Klein (Lyft) describing how Envoy solves the difficulties of running a modern distributed SoA architecture using a service mesh and Dinesh Dutt (Cumulus Networks) on troubleshooting data center networks using better methods than just old traceroute and ping.
- Orchestration, Scheduling, & Containers ( 7+ sessions): includes Seth Vargo (HashiCorp) on using Vault for microservices password/credentials management and Dharmesh Kakadia (Microsoft) on how orchestration scheduling works in Mesos, Kubernetes, and Docker Swarm.
- Distributed Data & Databases (6+ sessions): includes Miles Ward (Google) showing how the Google Cloud Spanner differs from other DBMS approaches and Avantika Mathur (Electric Cloud) on the best strategies for database deployments and rollbacks, and the best practices for reliably deploying databases as part of a CD pipeline.
- Systems Engineering (6+ sessions): includes Bryan Liles' (Capital One) tutorial on distributed tracing and Tammy Butow's (Dropbox) boot camp on Chaos engineering.
- Hardware, Storage, & Capacity Planning (5+ sessions): includes Jack Chan's (Shutterfly) tale of how to migrate metadata from over 10B photos from a private datacenter into AWS in 100 days, and Patrick Reynolds (GitHub) on using Spokes, a custom, application-level replication system, to provide redundancy and scalable capacity for the Git service.
- Resilience Engineering (5+ sessions): includes Nora Jones (Google) and Aaron Blohowiak (Google) leading separate talks on chaos engineering, purposefully injecting failure, and the precision chaos techniques that combat service failure.
- Security (5+ sessions): includes James Wickett (Signal Sciences) on building and securing a complete serverless application, plus sessions on the Zero Trust security model and the edge threat denying Incident Command process at Fastly.
- Distributed Systems (4+ sessions): includes Henry Robinson (Cloudera) on the best ways to build/scale distributed systems and Caitie McCaffrey (Twitter) on the top strategies for testing/verifying distributed systems.
- Technical Leadership (5+ sessions): includes Julia Grace's (Slack) explorations of the most effective ways to determine if your teams are truly functioning on a high level and Kathleen Vignos' (Twitter) solutions for stabilizing engineering teams in work environments where constant change and re-organization is the norm.
Want to keep your team updated by Velocity's best and brightest? Then get this video compilation and remember—you can explore everything at your own pace by downloading the video.