re:Invent 2020: Open source session round-up

Our previous post, re:Invent 2020: Open Source track preview, rounded up sessions in our official re:Invent Open Source track. In this post, we’ve collected additional open source-related content that is spread across other re:Invent tracks during the three-weeks of the virtual event. A variety of open source sessions are spread across re:Invent talk tracks, covering topics such as machine learning, compute, mobile and web applications, and developer tooling. We also have AWS partner and customer sessions that are well worth attending, so make sure you check those out, too.

Compute

HPC on AWS: Innovating without infrastructure constraints—Ian Colle
Colle will be covering how to migrate High Performance Computing (HPC) workloads to AWS to accelerate your most complex engineering simulations and high-performance data analysis workloads. Attendees will learn about the open source technologies AWS ParallelCluster and Amazon FSx for Lustre used, as well as how to increase agility and accelerate life-saving research in the development of drugs and vaccines.

Containers

Getting an insight into your Kubernetes applications—Sudeeptha Jothiprakash
Jothiprakash will explain how to combine open source tools, such as Prometheus and Grafana, with Amazon CloudWatch using CloudWatch Container Insights. Attend to this session for a demo of Prometheus metrics with Container Insights.

Define AWS service resources with AWS Controllers for Kubernetes—Jay Pipes
This session covers one of the many open source projects for customers using Kubernetes: AWS Controllers for Kubernetes. Until now, if you had dependencies on an AWS-managed service resource—an Amazon S3 bucket, an Amazon SNS topic, an Amazon DynamoDB table, and so on—you needed to use a tool such as Terraform or AWS CloudFormation to manage the creation and life cycle of those resource dependencies. With AWS Controllers for Kubernetes (ACK), you can now define your application’s AWS-managed service resources using Kubernetes API and manifests. There is no need to use a different configuration system or log in to the AWS console. Session attendees will learn about the design of the ACK, the features provided, and the roadmap for service integration.

Databases

Running Apache Cassandra workloads with Amazon Keyspaces—Arturo Hinojosa and Mihir Desai
In this session, learn about running and securing mission-critical Apache Cassandra workloads and how customer PwC is using Amazon Keyspaces (for Apache Cassandra) as part of a solution to store data for its proximity contact-tracing mobile application. This session will cover architecture, key features and functionality, as well as how to design and visualize Amazon Keyspaces (for Apache Cassandra) data models using the NoSQL Workbench.

Developer tools

AWS CDK: What’s new and what’s next—Jason Fulghum
The AWS Cloud Development Kit (AWS CDK) is an open source project that allows you to define cloud infrastructure and reusable components in “real code” and provision through AWS CloudFormation. Session attendees will learn about what the AWS CDK has been up to this past year, including new feature launches, new related project releases (such as CDK for Kubernetes [cdk8s] and CDK for Terraform), and community growth. The lead development manager for AWS CDK will dive into interesting changes from the past year, and look ahead to what’s next for AWS CDK and where it’s headed in 2021.

AWS Copilot: Simplifying container development—Efe Karakus
AWS Copilot is an open source command-line tool that makes building, developing, and operating containerized applications on AWS a breeze. In this session, learn how AWS Copilot can help you and your team manage services and safely deploy them to production.

Internet of Things

Developing and deploying modern edge applications at scale—Richard Barry
Learn about the latest capabilities of AWS IoT device software, and find out how customers have developed and deployed edge applications that manage fleets comprising millions of devices. Building a successful IoT solution depends on the potential of connecting devices at scale while they’re sitting at the edge in our homes and offices, in factories and oil fields and agricultural fields, in planes and ships and automobiles—everywhere. We’ll hear how FreeRTOS and AWS IoT Greengrass can help manage these devices, take action, and aggregate data.

Machine learning

Fast distributed training and near-linear scaling with PyTorch on AWS—Aditya Bindala
Efficiently scaling PyTorch training jobs to hundreds or thousands of GPUs is challenging. This session presents the most common issues and bottlenecks data scientists encounter with multi-node training. You will find out how AWS helps solve these challenges and removes network bottlenecks with optimizations to PyTorch and AllReduce algorithms as well as how to reduce training time and lower costs while maintaining high scaling efficiency.

Train large models with billions of parameters in TensorFlow 2.0—Can Karakus
When training with large models or large inputs, a single GPU’s memory is often a limiting factor during training, producing out-of-memory (OOM) errors. The solution to this problem is model parallelism, which places layers or operations of a model on multiple GPU devices to utilize the aggregate memory of all GPUs in the cluster. But implementing model parallelism efficiently is hard, time-consuming, and requires expertise in graph partitioning and setting up execution pipelines. In this session, learn how to train large models with billions of parameters using Amazon SageMaker and TensorFlow 2.0.

Mobile and web applications

Power modern serverless applications with GraphQL and AWS AppSync—Ed Lima
Lima will share how AWS AppSync and GraphQL can help you implement powerful, secure, highly available, flexible, resilient, and scalable backends for all types of applications with increased development velocity. GraphQL is a technology that improves application performance and enables businesses to build applications faster. AWS AppSync is a fully managed service that lets you create scalable GraphQL APIs to securely access and combine data from multiple sources.

Empower front-end web and mobile app development with AWS Amplify—Richard Threlkeld and Mohit Srivastava
Find out about the fast-changing trends of web and mobile application development that customers are using to deliver superior customer experiences. Threlkeld and Srivastava will explain how they are using AWS Amplify for development, and AWS Amplify Console for deployment and hosting. They also will review the Amplify libraries, UI components, CLI toolchain, and more.

Build iOS & Android mobile apps in record time with Flutter and AWS Amplify—Michael Labieniec
Learn how to build Flutter apps with AWS Amplify’s deeply integrated programming libraries and CLI. Flutter is Google’s UI toolkit for building natively compiled applications for mobile, web, and desktop from a single code base. Harness the speed and performance of building with Flutter and seamlessly connect to a cloud backend. With Amplify Flutter, you can quickly create, configure, and integrate categories like Auth, Storage, and DataStore to build feature-rich, full-stack applications for use cases such as signing in and signing up users, uploading and downloading files, and offline and online scenarios.

Additional sessions covering AWS Amplify and AWS AppSync to check out include Unify access to siloed data with AWS AppSync GraphQL resolvers presented by Nader Dabit, and Best practices to securely operate GraphQL at scale with AWS AppSync presented by Brice Pelle

AWS Partner sessions

Red Hat

Red Hat OpenShift Service on AWS
Learn about Red Hat OpenShift Service on AWS and the capabilities it provides. Find out how you can leverage APIs and existing Red Hat OpenShift tools for deployments on AWS, and get a demo of the service. This session will have a walk through of the technical implementation, and share lessons learned during initial service development.

Customer sessions

Running ultra-high throughput ad tech workloads on Aerospike
In this session for technology executives, architects, and engineers, dive deep into the AWS high-performance playbook for running ad tech workloads in the cloud using Aerospike, an open source low-latency NoSQL database platform. The session covers how The Trade Desk implements Aerospike on AWS to support millions of queries per second at the edge for real-time bidding and peak loads of 20 million writes per second in its cold storage of user profiles. Learn best practices for configuring Aerospike on AWS for ad tech workloads, including recommendations for data engineering, architecture, and optimal Amazon EC2 instance types for performance and cost efficiency.

Scaling MLOps on Kubernetes with Amazon SageMaker Operators
Amazon SageMaker Operators for Kubernetes is an open source project that provides operators that can be used to train, optimize, and manage machine learning models as well as set up inference endpoints in Amazon SageMaker from their Kubernetes cluster. Session attendees will learn how Intuit built a Kubernetes controller that wraps around operators using SageMaker, Apache Spark, and Argo to run world-class, high-availability systems.

Productionizing R workloads using Amazon SageMaker, featuring Siemens
Learn how Siemens Energy—a technology provider for more than one-sixth of the global electricity generation, with more than 91,000 employees and presence in more than 90 countries—is enabling new digital products with Amazon SageMaker to build, train, and deploy statistical and ML models in R at scale. R language and its 16,000+ packages dedicated to statistics and machine learning are used by statisticians and data scientists in industries such as energy, healthcare, life science, and financial services. Using R, you can run simulations and ML securely and at scale with Amazon SageMaker, while reducing the cost of development by using the fully elastic resources in the cloud.

Simplifying delivery as code with Spinnaker and Kubernetes
Spinnaker is an open source continuous delivery platform for releasing software changes with high velocity and confidence. In this session, Netflix and AWS discuss their collaboration on Spinnaker’s next-gen vision for transitioning from an imperative mix of pipelines and stages to a more declarative description of end goals and bring the “delightful” back to “delivery.” You might also want to check out session Going global in minutes: Multi-Region expansion simplified. In that session, Autodesk will share how they are using this open source project.

Join us

re:Invent is a free 3-week virtual conference that will be held November 30 — December 18, 2020. Free registration is now open to all. Register now to view the agenda and add sessions to your calendar.