This presentation will take you through the biggest areas where you need to focus your efforts in order to keep your data safe at AWS, and will show some real-life examples of what could go wrong if you make compromises or allow bad practices.
Amazon uses a “pay as you go” pricing model: you pay for the resources that you use, and in most cases don’t need to pre-allocate resources. While this allows your business to scale, it means that each component of your data pipeline will incur a separate charge, which can obscure the overall cost of running the pipeline. This talk will examine those changes, along with strategies for partitioning those costs between your clients or organizational units.
Abstract Since last ETE, we’ve seen two more major versions of the JDK! What’s changed in the last year, and what’s coming down the road? Join Java Language Architect Brian Goetz in a whirlwind tour of just some of the features coming to Java next. About Brian Goetz Brian Goetz is the Java Language Architect at Oracle, and was the specification lead for JSR-335 (Lambda Expressions for the Java Programming Language.) He is the author of the best-selling Java Concurrency…
Abstract The Go programming language emphasizes simplicity and speed. Common programming mistakes are detected by the compiler. The language itself encourages proper error handling. It has first-class concurrency support using goroutines and channels. And on top of all this, Go offers lightning-fast compilation and execution. This talk will cover all the unique features that could make Go the ideal choice for your next project! About Jay McGavren Jay is the author of Head First Ruby and Head First Go (both…
Abstract Have you heard about the latest garbage collectors (GCs) in OpenJDK HotSpot VM? Are you interested in learning more about their algorithm and design goals? Would you like a basic comparison between them and the current default collector in OpenJDK? If you answered yes to any of the above questions, then this is the session for you! In this session Monica Beckwith will talk about ZGC, Shenandoah GC and compare them to G1 GC (the current default GC). About…
Abstract At FB, we run containers with Linux, Btrfs and systemd. But how does this approach work at scale, and what is the philosophy behind this approach? In this talk, I will dive into the details of these components, including: Advanced details of btrfs and the benefits of a composable file system on a high-traffic network How we use systemd on the host and inside a container New Linux kernel features such as cgroupv2 and bpf About Lindsay Salisbury Lindsay…
Abstract Great automation doesn’t replace humans; it enhances us. When we are choosing or building tools for our team, we want them to play like team members: keep us informed, make the easy decisions repeatably, and pass the hard decisions to the humans along with the information we need to make them. Based on research in human-centered design, this talk enumerates principles and challenges of collaboration for programs. It lists strategies for eliminating “human error” as well as human frustration….
Abstract What is the Internet of Things? Does adding the internet to a device (or thing) make it better or does it just create more problems? What has IoT done well? What are some problems with IoT? What can be done better? About Don Coleman Don is a software developer who enjoys hacking with phones, Arduino, and other hardware. He works as the CIO of Chariot Solutions, and is an adjunct professor at NYU. Don is the co-author of Make:Bluetooth and Beginning NFC.
Abstract So you want to connect a “thing”…. or many “things”? In this journey, you need to make a very important decision (among many): which wireless protocol is right for your use case and how do you choose your platform? At the end of the day, no matter how good your application, if you don’t have reliable connectivity, you just have a [dumb] thing! In this talk, we’ll cover the details of LoRa, how it works, what makes it special,…
Abstract Functions are the key superpower enabling serverless architectures. Certainly the low cost of only paying for what you use (100% utilization) is a nice benefit. And no doubt there are operational savings by only focusing on your business logic instead of managing a brittle and complicated infrastructure. But designing with functions as the smallest unit of deployment also confers a much less talked about hidden superpower: functions are faster to deploy! Deploying a single function usually completes within 1…
