Track: Sponsored Solution Track IV

Agile to DevOps. A Better Way?

Together we are on a share quest; to make our customers happier, our organizations more successful, the world a more connected place. Our tool of choice is software. Our goal to delivery more compelling software, faster! Our go to answer is Agile and evolving to DevOps. But what does this really mean for the existing enterprise and how do we get there?

In this session, we will explore these questions and potential answers through the examination of a past development project gone away.

Making Session Stores More Intelligent

In this session, we will explore how to extend session stores beyond just persisting authentication. By centralizing the session state (potentially de-coupled from user records) in a database it is possible to provide user experience enhancements, deep analytics, notifications, and personalization. Techniques examined in this talk will be using session-linked Bloom filters to surface fresh content, site-wide or group-based notifications and bitfield-based activity pattern monitoring.

Demos and code in this session will be in Node.js and Redis, although the patterns and techniques are widely applicable to any HTTP-based platform or stack. The techniques will be illustrated in an e-commerce example environment, but any user-base environment could benefit (CMS, social network, gaming, etc). While this talk examines the more advanced sides of session storage, the approach aims to be both practical and approachable.

Multi-host, Multi-network Persistent Containers

Containers are great vessels for your application’s ephemeral data, but what about the data that drives your business? It must survive containers coming and going, maintain its availability and reliability, and grow when you need it. In this talk, we will discuss strategies for working with persistent containers, where you can store your data, and how to scale your persistent container layer. We will include code samples and interactive demos showing the power of Docker Machine, Engine, Swarm, and Compose, combined with multi-host networking, to build a reliable, scalable, and production-ready tier for the data needs of your organization.

Building Real World Node.JS Microservices on Azure

Microservces are small services with independent lifecycles that work together” -Sam Newman. In this talk we’ll explore building real world microservices application using using Node.JS and kubernetes on Azure. After a brief introduction of what microservices are and why they are important, we'll spend the bulk of the time looking at how BDD, CQRS, and Event Sourcing coupled with modern CI/CD techniques can help you deploy containers to Azure. We’ll explore best practices, code, and tooling to get the job done right. Beginners will get a sense of what microservices are and what makes different, whereas more experienced practitioners will get an insight into practical advice into how to implement them.

Human in the Loop AI

Machine Learning applications need to continually update their models with new training data to improve and maintain accuracy. However, it is often difficult to decide what new data needs to be labeled for training, and what are the best workflow and interfaces for labeling. This talk will focus on how you can use Active Learning to improve your training data at scale with common Deep Learning frameworks. At the end of this talk, you will understand several Active Learning strategies that you can apply for your business needs. We will use the example of applying Active Learning to the ImageNet data set using the TensorFlow Deep Learning framework.

Challenges of Distributing Postgres: A Citus Story

Set theory forms the basis for relational algebra and relational databases, and SQL is the lingua franca of modern RDBMS’s. Even with all the attention given to NoSQL in recent years, the lion share of database usage remains relational. But until recently, nearly all relational database solutions have been limited to the resources of a single node. Not anymore.

This talk is about my team’s journey tackling the challenges of distributing SQL. Specifically in the context of my favorite (open source) database: Postgres. I believe that too many developers spend too much time worrying about scaling their databases. So at Citus Data, we created an extension to Postgres that enables developers to scale out compute, memory, and storage by distributing queries across a cluster of nodes.

This talk describes the distributed systems challenges we faced at Citus in scaling out Postgres—and how we addressed them. I’ll talk about how we use PostgreSQL’s extension APIs to parallelize queries in a distributed cluster. I’ll cover the architecture of a distributed query planner and specifically how the join order planner has to choose between broadcast, co-located, and repartition joins in order to minimize network I/O. And if there’s time, I’ll walk through the dynamic executor logic that we built. The end result: a distributed database and a lot less time spent worrying about scale.