Log in to bookmark your favorites and sync them to your phone or calendar.

PRO: Scalability: DevOps & Performance & Digital Transformation [clear filter]
Thursday, June 20

1:00pm EDT

PRO TALK: Functional Programming Fundamentals
This workshop aims to be the entry point for developers into the world of functional programming. We'll talk about various FP paradigms such as Referential Transparency, Immutability, Higher Order Functions etc. Examples and hands on training will be via the Clojure programming language. After learning about the fundamentals of FP concepts (and getting our hands "dirty" with some Clojure code), we will progress to modeling and building a simple web app. We'll start small and show how FP principles lend themselves to our solution. Depending on how much time there is for workshops this can be very short intro or a much longer fundamentals course.

avatar for Ronen Cohen

Ronen Cohen

Tech Lead, AppsFlyer
Ronen Cohen is a software engineer who enjoys architecting systems for scale. He is passionate about everything software: from OS internals to distributed computing, weird data-structures to functional programming. Ronen is a tech lead at AppsFlyer, helping build systems to process... Read More →

Thursday June 20, 2019 1:00pm - 1:25pm EDT
Workshop Stage 1

1:30pm EDT

PRO TALK: Lessons Learned in Building CI/CD for Serverless Architectures
https://slides.com/anthonyc-1/lessons-learned-in-building-ci-cd-for-serverless-architectures/live#/ Why Serverless architecture (SA)? What are its pros (and some cons)? Common Pros: - Lower CapEx/OpEx Common Cons: - Increase complexity Key-enabler of other paradigms - CI/CD - Event/Data-driven - Agile (which affects business units; technology enabling business) Why we need CI/CD (for SA)? Fast iteration/Increase productivity/Time to Market/Lower cost -> Business win Why Serverless architecture (SA) as a paradigm is ideally suited for integration with CI/CD? How What are some of the general guideline and architecture principle for SA to make suitable for CI/CD? General principals of microservice (MS) all apply to SA, but it’s even more suitable as with SA you no longer manage your infrastructure the same way. What are some of these principals for MS that applies to SA? * Your SA infrastructure (as code) can now more closely be treated as code. You can and should treat your SA part of the infrastructure as cattle. ** THIS ENABLES CI/CD. ** Small, single unit of responsibility: This is a core tenant of UNIX and computing, do one thing and do it well. This leads to smaller blast radius; easier to unit test and refractor. When you build SA, you follow this same principle. ** THIS ENABLES CI/CD. ** Short-lived and stateless: As characteristics of Serverless Functions; this forces the developers to think about clearly delineating computing functions and its data. ** THIS ENABLES CI/CD and also design data driven architecture ** What are some its pitfalls? (See Challenges) What kind of tools may be helpful? (See part of How) Your tooling should be flexible but manageable. I recommend using SA framework (we use Serverless) to help build your SA infrastructure. For CI/CD, I would look at your organization security requirement and see if your CI/CD Challenges Greater technological complexity; can be difficult to troubleshoot. This can be alleviated by building your SA architecture into units of functions; for example we use step functions to manage lambda functions so that we have a clear way of how the lambda functions interact. Cultural: Your infrastructure team need to understand core tenants of DevOps, understand SA and CI/CD. Ideally your team should already have experiences of some of IaC, SA, CI/CD. There are some required skillsets to achieve these goals. Security: Restricting Production from developer; requires every deploy from CI/CD Newer tools that are less mature. Your need to treat your SA infrastructure as code. How we overcame those challenges (Show) How our Data Engineering team (intersection.com) leverages Serverless Framework to implement our data pipelines with CI/CD. I will give a overview and demo on how we use Serverless Framework to create Data Pipeline with AWS Step Function, all of which are part of a CI/CD pipeline. Recommended: Start small; start SA first before CI/CD and/or start CI/CD without SA.

avatar for Anthony Cheng

Anthony Cheng

Data Engineer, Intersection
Data/DevOps Engineer with more than a decade of infrastructure experiences, my career has spanned from working on "Iron Age" data center with network and system infrastructure to working in programmatic "Infrastructure as Code" with AWS. In my career, I have worked with start-up... Read More →

Thursday June 20, 2019 1:30pm - 1:55pm EDT
Workshop Stage 1

2:00pm EDT

PRO TALK: The Future of Applications: Building Real-Time Streaming Apps That Transcend Devices and Clouds
This session presents a new model for building real-time applications using stateful Web Agents and multiplexed streaming APIs. Using the open source Swim platform, developers can quickly and easily build stateful, decentralized applications and real-time user experiences. Swim applications automatically optimize usage of available cloud, edge and network resources to ensure streaming updates are continuously delivered in real-time. Come see how easy it is to build ultra-low latency, massively distributed streaming applications without ever depending on a database, message broker or app server. The future of applications is collaborative, real-time and stateful. If you're building next-generation software (for AR/VR, autonomous vehicles, artificial intelligence, smart cities, etc), the biggest challenge remains in efficiently processing and distributing multiple real-time data streams. Furthermore, it's critical to be able to combine data streams to identify correlations and power real-time user experiences. So why are we still using stateless, complex, slow, request-driven REST architectures?

avatar for Rohit Bose

Rohit Bose

Senior Software Engineer, SWIM.AI
Rohit Bose is a senior backend engineer for Swim.ai. Having joined Swim.ai in 2016, Rohit works with all layers of the technical stack, from operating systems to web clients. He appreciates clever algorithms and integrations, and has nightmares about pixels and UI animations. In his... Read More →

Thursday June 20, 2019 2:00pm - 2:25pm EDT
Workshop Stage 1

2:30pm EDT

PRO TALK: Open Source Governance at Massive Scale
It is one thing to put a project up on GitHub and call it "Open Source." It is quite another to commit to growing and governing a project developed and maintained by contributors across the world. Both technical and community skills must be practiced simultaneously - a project cannot be successful with only one or the other. Open Source governance is hard enough with one or two projects. Now imagine trying to consistently and fairly govern a sprawl of 1200+ (and growing!) Open Source projects with complex technical and human interdependencies. Automation certainly helps with testing and releases, but it's also critical to maintain the personal touch that ensures both higher quality software and makes working in Open Source so rewarding. Come to this talk to hear a tale of a medium-sized startup experiencing the pain of Open Source sprawl and how they mitigated that pain by developing consistent governance so that their Open Source community not only survived but is thriving.

avatar for Nell Shamrell-Harrington

Nell Shamrell-Harrington

Principal Software Development Engineer, Chef Software
Nell Shamrell-Harrington is a principal software development engineer - community engineering lead and member of the Habitat core team at Chef. Additionally, she is a member of the Board of Directors of Operation Code, a non-profit dedicated to teaching military and veteran’s software... Read More →

Thursday June 20, 2019 2:30pm - 2:55pm EDT
Workshop Stage 1

3:00pm EDT

PRO TALK: Trace-Driven Development
Distributed tracing is invaluable to operating software, allowing operators and developers alike to quickly diagnose production issues. So why do we keep treating it like a "day two" problem, stashing it away as yet another TODO for once things are running and all the fires are put out? In this talk, I'll demonstrate how you can apply distributed tracing during the earliest parts of the software development process so you can set -and validate - expectations about your applications from the get go. This includes early identification of possible performance hotspots, compliance with other instrumented RPCs, and visualization of the shape of a request as it traverses your system. Expect to walk away wanting to instrument all your code ASAP. Remember: only YOU can prevent codefires! You shouldn't need any background in tracing or other instrumentation tools, as I'll explain the relevant concepts and terminology. Audience members of all technical levels are welcome.

avatar for Austin Parker

Austin Parker

OSS Engineer, LightStep
Austin Parker has been solving - and creating - problems with computers and technology for most of his life. He is an Open Source Software Engineer at LightStep and maintainer of the OpenTracing project in an effort to help other developers write, and operate, more reliable software... Read More →

Thursday June 20, 2019 3:00pm - 3:25pm EDT
Workshop Stage 1

3:30pm EDT

PRO TALK: DevOps: How to Be an Ansible Contributor
Contributing to Open Source projects like Ansible can be a daunting and perhaps scary endeavor. Even though most Open Source projects have getting started guides and tutorials, there are always those little bits that are missing. A missing bit might be the crucial element that makes the difference between you contributing or not. Books can help, podcasts are good, videos almost get you there, but nothing is better than a developer's step-by-step walk-though of their own personal Ansible contribution experience. This session will take you through how and why I contributed to Ansible and how you can. Arming you with what you need to know, things like Python, git, and YAML. Going over how to set up your development environment and contribute your code. Then wrapping up with what happens when you submit your contribution, how to avoid common mistakes, and your contribution becoming part of an Ansible release.

avatar for John McDonough

John McDonough

Developer Evangelist, Cisco DevNet
John McDonough has been programming since he was in 6th grade. Now several decades out of college, he’s been around awhile. John specializes in infrastructure automation. Whatever OS you want to run, whatever App you want, any Service you want … John can show you how to automate... Read More →

Thursday June 20, 2019 3:30pm - 3:55pm EDT
Workshop Stage 1