The annual Microsoft Build is finally here, bringing a lot of new technology and feature updates. I don’t know if you guys stayed up late like me and watched the whole process of the opening ceremony and the five core themes? Next, let me talk to you about the seven major technical updates that developers should pay attention to most.
Improvements in AI capabilities
▌ Azure OpenAI Service is finally here
Developers can use Azure OpenAI Service to complete contextual reasoning, reading comprehension, and related application scenarios for writing assistance, code generation, and understanding unstructured data. Using Azure OpenAI Service, different models can be set based on application scenarios, including GPT-3 series ( Ada/Babbage/Curie/DaVince ) and Codex series ( Davinci-codex / Cushman-codex ). Below I will summarize the capabilities of Azure OpenAI Service based on GPT-3 and Codex related models, hoping to help you in your application.
▌ Improvement of Azure Machine Learning Service Capability
Azure Machine Learning is a cloud service for accelerating and managing the lifecycle of machine learning projects. Machine learning professionals, data scientists, and engineers can use it in their day-to-day workflows: training and deploying models, and managing MLOps. At Build 2022, Microsoft brought a brand new Azure ML CLI (v2), and machine learning managers can participate in managing the machine learning lifecycle through a new command line. You can train machine learning models using assets such as compute resources, environments, and datasets through the Azure ML CLI(v2).
E.g:
Create compute instances directly via the Azure ML CLI (v2)
az ml compute create --name "testdev-vm" --size STANDARD_DS11_V2 --type ComputeInstanceCreate compute clusters directly through the Azure ML CLI (v2)
az ml compute create --name "aml-cluster" --size STANDARD_DS11_V2 --max-instances 2 --type AmlComputeAnd the AutoML feature that everyone is concerned about has also added support for natural language processing. Can be seamlessly integrated with the Azure Machine Learning Data Labeling feature to label textual data or ingest existing labelled data.
▌ Integration and Improvement of Cognitive Services Functions
Of course, there are also enhancements to traditional Cognitive Services, such as the layout of paragraphs, headings, and headings in Azure Form Recognition for more precise text extraction, Azure Bot Service and Power Virtual Agents integration, and Azure Metrics Advisor's new auto-tuning feature enables users to customize the service to display only personalized anomalies.
.NET full-scene development has finally arrived
MAUI is finally officially released, which completes all the features of .NET 6. You can use C# to complete cross-platform application development through MAUI. In addition to the traditional native development support, you can also combine Blazor Hybrid to complete related application development.
In addition to the official release of MAUI, there is also the support of development tools. In addition to the MAUI support of Visual Studio 2022, Visual Studio 2022 for Mac is also officially released to support the development of MAUI. It is a pity that the support of the official version of Visual Studio 2022 for Mac in MAUI is not perfect at this stage. It is recommended that all macOS friends consider using Visual Studio 2022 preview for Mac (17.3 Preview 1) / Visual Studio Code (Comet) for related work development.
Of course, while .NET 6 is improving, .NET 7 is also previewing. Like the Minimal API that everyone is concerned about, WASI support will continue to be supported in subsequent versions.
Contributing to the cloud-native era
▌Azure Container Apps is finally here
The general availability of Azure Container Apps, which can execute application code packaged in any container, is not affected by runtime or programming model.
Advantages of Azure Container Apps:
- Forget about managing cloud infrastructure and complex container orchestrators.
- Serverless support
- Extend HTTP requests, events or run always-on background jobs
- Automatic encryption of ingress and service-to-service communication
- Based on AKS, KEDA, Dapr and Envoy, you can complete a very good cloud-native solution through Azure Container Apps and Dapr. The following is an e-commerce architecture solution based on Dapr + Azure Container Apps + Azure Service.
▌Draft enhances the development experience of AKS
We know that Draft ( https://github.com/azure/draft ) enables developers to take non-containerized applications and generate Dockerfiles, Kubernetes manifests, Helm charts, Kustomize configurations, and other components related to containerized applications It's easier to start building applications that run on Kubernetes. Draft can also generate GitHub Action workflow files to quickly build and deploy applications to any Kubernetes cluster.
The new Draft released on Microsoft Build 2022 has been integrated into the AKS CLI, you can install the AKS-Draft plugin, and developers can quickly create, containerize and deploy related applications to Azure Kubernetes Services through Draft.
▌ Other functional updates In addition to Azure Container Apps, AKS and other related content, Microsoft also provides gRPC support for Azure App Services, which can host gRPC applications on Azure App Services. There are also enterprise-oriented Azure Spring Apps, which adds Service Bus Explorer, etc.
Teams development function is gradually improved
The Teams toolkit plugin for Microsoft Visual Studio Code helps create and deploy Teams apps with integrated identity, cloud storage access, data from Microsoft Graph, and other Azure and Microsoft 365 services with a zero-configuration approach. For Teams app development, the Teams toolkit allows creating, debugging, and deploying Teams apps directly from Visual Studio Code.
The new Teams SDK enables developers to build Live Share capabilities into their applications, enhancing the real-time collaboration experience in Teams meetings. The Live Share Experience Tool is built on the Microsoft Fluid Framework and allows developers to create synchronized meeting experiences with a few lines of code from the client.
Low-code to move on
▌ Power Virtual Agent new intelligent robot development experience
It combines the sophistication of the pro-code capabilities of Azure Bot Framework Composer with the simplicity of the Power Virtual Agent low-code platform. This will make it easier for professional developers and business experts to collaborate in one Microsoft Intelligent Robotics Studio.
▌Power Pages, a new member of PowerPlatform
Microsoft Power Pages is a low-code development and hosting platform ideal for building business-focused websites, allowing low-code makers and professional developers to design, configure, and publish websites for desktop and mobile devices with a smooth visual experience. Previously, Power Pages existed as a feature in Power Apps called the Power Apps Portal. Power Pages will be the fifth member of the Power Platform family as a standalone product within the Power Platform portfolio. The new Power Pages brings a new design studio, a very rich template center, and integration with Visual Studio, GitHub and Azure DevOps through Pro-Developer to help developers achieve advanced business needs. Of course Power Pages also opens up the Microsoft Dataverse, allowing users to securely collect business information and share it with site visitors.
▌ Unattended Robotic Process Automation (RPA) and Virtual Machines (VMs) in Power Automate
By hosting Azure-powered VMs such as Windows 365, organizations can use Power Automate's built-in autoscaling capabilities to expand VM capacity, enabling automatic RPA flows up or down as needed.
Changes in the Windows development ecosystem
▌ The ARM ecosystem is here
Microsoft is working on a new Windows-on-Arm development kit powered by the Qualcomm Snapdragon Compute Platform that will allow Windows developers to build Arm-native applications.
Based on Project Volterra, Microsoft has launched a comprehensive ARM-native developer toolset, which includes ARM-native versions of Visual Studio 2022, Visual Studio Code, .NET, and the .NET Framework.
Microsoft is also working on projects like Open JDK, Python, LLVM, Node, and Git to support ARM64.
Project Volterra also enables developers to leverage the powerful integrated neural processing unit (NPU) to build applications that execute native AI-accelerated workloads. Windows-based Project Volterra dynamically transfers model AI application inference load between client and cloud through Hybrid Loop mode.
▌ More powerful Microsoft Edge WebView2
Microsoft Edge WebView2 is Microsoft's latest web framework for building embedded browsing experiences. WebView2 can be used with the .NET MAUI framework to help developers embed web-based experiences. WebView2 for WinUI 2/UWP is also coming soon.
We know that Internet Explorer 11 (IE11) will be retired on June 15, 2022. However, Microsoft Edge provides backward compatibility with Internet Explorer mode for users and organizations that need to continue to access critical Internet Explorer-based web applications. For web developers, this means that there is no longer a need to work on backwards compatibility with IE11. For application developers who embed web content via the MSHTML (Trident) engine, the application will remain unchanged after IE11 is retired and will continue to be supported.
▌ Preferred platform for full stack developers - Windows 11
Windows 11 is creating an environment that developers like. Developers can develop artificial intelligence, Android, cloud native, IoT web pages and other related applications through Visual Studio / Visual Studio Code combined with WSL / WSA / Windows Terminal / Microsoft Edge. As an open platform, Windows welcomes developers and applications across platforms and any programming language, including .NET, Web, Android, C++ and Linux. Developers can design applications for Windows using their usual technologies and development tools.
A more intelligent big data platform
▌ Microsoft Intelligent Data Platform released
Introducing the Microsoft Intelligent Data Platform, a new integrated platform that unifies database, analytics and governance, enabling organizations to spend more time creating value than integrating and managing fragmented data assets.
The Microsoft Intelligent Data Platform enables organizations to easily adapt and adapt in rapidly evolving environments, add layers of intelligence to their applications, unlock predictive insights and manage their data from anywhere. It addresses major customer pain points such as market fragmentation, while also enabling customers to capture and leverage data more effectively to increase market share.
▌ SQL Server 2022 Release Preview
SQL Server 2022 is the most Azure-ready version of SQL Server with improvements in performance, security, and usability.
SQL Server 2022 integrates with Azure Synapse Link and Microsoft Purview to enable deeper analysis, prediction and governance from large-scale data. Enhanced cloud integration with disaster recovery (DR) to Azure SQL Managed Instance and ETL-free (extract, transform and load) connectivity to cloud analytics, which enables database administrators to manage with greater flexibility and minimal impact their data assets.
▌ Azure Cosmos DB development is more friendly
The new capabilities of Azure Cosmos DB enable developers to build scalable, cost-effective cloud-native applications and add enterprise-grade functionality to their applications. Developers also have new, free ways to test online and offline. Azure Cosmos DB is now available with a 30-day free trial and supports MongoDB data in the Azure Cosmos DB Linux desktop emulator.
▌ Azure Synapse Analytics feature update
Azure Synapse Link for SQL enables near real-time insights by removing barriers between operational data stores and Azure Synapse Analytics. For SQL Server 2022 and Azure SQL Database, Azure Synapse Link acts as a seamless data pipeline for Azure Synapse Analytics, so customers don't need to manage ETL logic. This makes it easier to do business intelligence (BI) reporting on operational data in near real-time.
Microsoft Graph Data Connect enables Microsoft 365 customers to easily leverage their ability to organize data, move it into Azure Synapse, and discover new actionable business insights that improve customer satisfaction, increase productivity, and optimize business processes. Microsoft 365 customers using Azure Synapse can use Azure Synapse Analytics to easily unlock new insights and leverage the valuable employee and customer collaboration data they already have.
postscript
The technical update of Microsoft Build 2022 is very large, covering cloud native, big data, Windows, artificial intelligence, low code, development technology and tools and other fields. As a developer, you can learn a lot of new skills and knowledge from it. The above parts are some new technologies and function updates that I think are more important and worth paying attention to. I suggest that you can also read the official documents to learn more.
Recommended Resources One-stop view of all Microsoft Build news
◆ More information and learning materials for different speech sessions of the conference
https://docs.microsoft.com/events/resources/build-2022-post/session-resources/?wt.mc_id=build22_subs_devrel
◆ Participate in more activities after Microsoft Build
Other resources
- To learn more about Azure OpenAI Service applications, please visit
https://blogs.microsoft.com/ai/azure-openai-service-helps-customers-accelerate-innovation-with-large-ai-models-microsoft-expands-availability/ - If you want to learn about Azure OpenAI Service, please visit
- https://azure.microsoft.com/en-us/services/cognitive-services/openai-service/?msclkid=4f323323cd6f11ec9ce43cc184aa6d38
- If you want to learn how to train a model in Azure Machine Learning Service using the Azure ML CLI(v2), visit
https://docs.microsoft.com/en-us/learn/paths/train-models-azure-machine-learning-cli-v2/ - Learn how to use AutoML to train natural language processing models in Python at
https://docs.microsoft.com/en-us/azure/machine-learning/how-to-auto-train-nlp-models - If you want to know more about the official version update of .NET MAUI, please visit
https://devblogs.microsoft.com/dotnet/introducing-dotnet-maui-one-codebase-many-platforms/ - To download the official version of Visual Studio 2022 for Mac, please visit
https://visualstudio.microsoft.com/thank-you-downloading-visual-studio-mac/?sku=communitymac&rel=17 - To download Visual Studio 2022 (Windows / macOS) 17.3 Preview, visit
https://aka.ms/vs2022preview - If you want to know more about .NET development in WASI, please visit
https://mybuild.microsoft.com/en-US/sessions/ddf48da7-d3e6-4ba5-affd-b5735ce205fa?source=sessions - For more updates on the .NET Minimal API, please visit
https://mybuild.microsoft.com/en-US/sessions/e6b8b078-2e41-47fd-a497-151f1b766e2b?source=sessions - To learn more about Azure Container Apps, please visit
https://azure.microsoft.com/en-us/services/container-apps/?ocid=AID3042118#overview - To learn more about developing AKS with Draft 2, visit
https://docs.microsoft.com/en-us/azure/aks/draft - If you want to learn about Microsoft Teams development, please visit
https://docs.microsoft.com/en-us/microsoftteams/platform/get-started/get-started-overview - If you want to learn more about low-code development, please visit
https://powerapps.microsoft.com/en-us/what-is-low-code/ - To learn more about the Microsoft Intelligent Data Platform, please visit
https://azure.microsoft.com/en-us/blog/introducing-the-microsoft-intelligent-data-platform/ - To learn more about SQL Server 2022, visit
https://www.microsoft.com/en-us/sql-server/sql-server-2022 - To learn more about Azure Synapse Analytics, please visit
https://docs.microsoft.com/en-us/azure/synapse-analytics/ - To learn more about the progress of Project Volterra for Windows on ARM, please visit:
https://blogs.windows.com/windowsdeveloper/2022/05/24/create-next-generation-experiences-atscale-with-windows/ - If you want to know more about Edge WebView2, please visit
https://docs.microsoft.com/en-us/microsoft-edge/webview2/
Long press to identify the QR code
**粗体** _斜体_ [链接](http://example.com) `代码` - 列表 > 引用。你还可以使用@来通知其他用户。