Exploring Dapr: Open-Source Runtime for Cloud and Edge with @marcduiker

()

GitHub Models

The speaker introduces GitHub Models, a platform for exploring and experimenting with AI models.
GitHub Models offers a curated collection of models with entitlements attached to GitHub accounts.
Users can interact with models through a playground, adjust parameters, and simulate their use in applications.
The speaker demonstrates how to switch between different models and access detailed information about them.
GitHub Models provides code integration, allowing users to start using models with code and access preconfigured development environments.
The speaker highlights the use of GitHub Models in the GitHub CLI, enabling developers to call AI models for tasks like generating questions or summarizing code.
The speaker concludes by encouraging viewers to join the waitlist for GitHub Models and explore the possibilities of building AI-powered applications.

Dapr: An Open-Source Runtime for Microservices

Dapr is an open-source project that has been in development for almost five years.
Dapr originated from Microsoft Research and was quickly moved to open source.
Dapr was submitted to the Cloud Native Computing Foundation (CNCF) in 2021 and is expected to graduate this year.
Dapr is designed for building distributed applications, particularly those using microservices.
Dapr addresses cross-cutting concerns in distributed applications, such as reliable communication, security, and state management.
Dapr runs as a sidecar process alongside applications, allowing it to be used with any programming language.
Dapr is an open-source runtime that can be used for cloud and edge applications.
Dapr can be used with applications that communicate via HTTP or gRPC.
Dapr offers client SDKs for various languages, including Node.js, Python, .NET, and Go.
Dapr provides a toolbox of APIs for microservice developers, including workflow, pub/sub, service invocation, and state management.
Dapr typically runs on top of Kubernetes, but can also be run on-premise or in major cloud providers.
Dapr uses a sidecar model, where a Dapr sidecar runs alongside each application service.
All communication between services and other resources, such as message brokers and storage systems, is done through the Dapr sidecar.
The sidecar model provides built-in resiliency, including retrying failed calls.
Dapr offers APIs for various functionalities, such as state management, pub/sub, and secret management.
Dapr itself is not a state store or message broker, but can integrate with hundreds of different resources.
Dapr's state management API can be configured to use various storage options, such as Azure Blob Storage or Redis.
Dapr is a flexible system that can be used for multi-cloud deployments.
Dapr allows developers to switch between local development environments and cloud environments with minimal code changes.
Dapr provides APIs that simplify common tasks, such as state management and pub/sub messaging.
Dapr can save developers time by allowing them to focus on domain logic rather than infrastructure concerns.
Dapr is often compared to service meshes, but they have different focuses.
Service meshes primarily deal with network routing and traffic management, while Dapr focuses on providing APIs for developers.
Both Dapr and service meshes can be deployed as sidecars.
Dapr is open-source and has a large community of developers.
Dapr provides quick start repositories with code samples in various languages.
Dapr offers both HTTP and client SDK APIs.

Dapr: State Management Example

The code snippet demonstrates saving state data using Dapr's State Management API.
The code defines a state object with a key and value, where the key is "order" and the value is the order itself.
The code uses the Dapr client to save the state to a state store, but the specific state store is not explicitly defined in the code.
The state store name is defined as "stateStore" and is mapped to a specific state store configuration in a separate file.
The Dapr CLI is used to run the application, specifying a configuration file that includes the state store configuration.
The configuration file defines the state store type, which in this case is a local Redis container.
The code saves 10 records to the Redis state store.
The example demonstrates switching state stores without modifying the application code by changing the configuration file extension.
The new configuration file points to an Azure DocumentDB state store, while maintaining the same metadata name "stateStore".
This allows the application to seamlessly switch between different state stores without code changes.

Dapr: Configuration and Security

Dapr can be used to switch out underlying resources without changing application code.
Dapr uses configuration files to define resources and secrets.
Dapr works with various secret stores, including local files and cloud-based secret stores.
Dapr uses Scopes to define security policies for services and resources.
Each Dapr microservice has a unique application ID.
Dapr uses YAML files for configuration, including resiliency policies.
Dapr provides default security and resiliency settings, but these can be overridden.
Dapr allows for flexible retry policies, including constant and exponential retry intervals.

Dapr: Community and Resources

Dapr is an open-source runtime for cloud and edge applications.
Dapr offers a variety of APIs for building microservices.
Dapr's documentation is comprehensive and detailed, covering various topics like building blocks and configurations.
Dapr has a YouTube channel with recordings of community calls, live streams, and user stories.
Dapr's YouTube channel also features recordings of longer sessions that delve into specific topics.
Dapr is designed to be gradually adopted, allowing developers to start with a few microservices and expand their use over time.
Dapr welcomes contributions and maintainers from the community.
Dapr has a variety of repositories for contributions, including the Dapr runtime itself, SDKs for different languages, and documentation.
Contributions to the Dapr documentation are highly encouraged, as they can improve the onboarding experience for new users.
Dapr has a dedicated Chinese translation of its documentation, but other language translations are welcome.
Before starting any work on a contribution, it is recommended to check if an issue already exists that matches the desired contribution. If not, create a new issue.
Dapr is an open-source project that allows backend developers to build microservices more efficiently.
Dapr is available worldwide and can be downloaded and installed anywhere.
Dapr can be run on top of Kubernetes, a Kubernetes-like environment, or locally.
Dapr has a large community and a Discord server for users to connect and collaborate.
The speaker encourages users to provide feedback on Dapr through LinkedIn, Twitter, or X.
The speaker is open to receiving both positive and negative feedback.
The speaker is excited to grow the Dapr community and improve the user experience.
The speaker is planning to do a hands-on Dapr live stream with Cecil, a Dapr community manager.