Building API in the cloud using Azure Functions

Building API in the cloud using Azure Functions.

Transcript

2. Building API in the cloud using Azure Functions Aleksandar Bozinovski Technical Lead, Seavus
3. PLATINUM GOLD SILVER BRONZE MEDIA PARTNER SUPPORTING PARTNERS OUR PARTNERS
4. Ве молиме исклучете ги мобилните уреди Please turn off your mobile devices
5. Agenda • Journey to Azure Functions • Developing Azure Functions in Azure Portal • Developing Azure Functions with Visual Studio • Hosting and pricing • Typical usage scenarios
6. App Service • Hosting applications inside of IIS (Internet Information Services) on Windows • Needs an App Service Plan • The App Service Plan defines the hardware resources • Made to support mainly web applications • Difficult to run • Batch processing • Scheduled tasks • Long-running processes
7. Web Jobs • Built-in triggers • Azure Storage Queue • Service Bus • Schedule tasks • Needs an App Service Plan • The App Service Plan defines the hardware resources • Difficult to scale
8. App Service Plan
9. Introducing Functions + Code Events + Data
10. Azure Functions • More triggers • Cosmos DB • Event Hub • HTTP • Web Hook • Pay-per-use • Doesn’t need an App Service Plan • Although can run on a App Service Plan too • Easy to scale
11. Azure Functions History App Service Functions Web Jobs
12. Azure Functions - Serverless Event-driven instant scale Micro-billing Abstraction of servers
13. Supported Languages
14. C# Script - CSX • Automatically referenced • mscorlib • System • System.Core • System.Xml • System.Net.Http • Microsoft.Azure.WebJobs • Microsoft.Azure.WebJobs.Host • Microsoft.Azure.WebJobs.Extensions • System.Web.Http • System.Net.Http.Formatting • Platform packages, use #r directly • Newtonsoft.Json • Microsoft.WindowsAzure.Storage • Microsoft.ServiceBus • Microsoft.AspNet.WebHooks.Receivers • Microsoft.AspNet.WebHooks.Common • Microsoft.Azure.NotificationHubs
15. Triggers and Bindings
16. DEMO 1 Create CSX Azure Function in Portal
17. Create Function App • Create Function App
18. Create Http Triggered Function • Create Http Triggered Function
19. Run Function • Run Function
20. Create Timer Triggered Function • Create Timer Triggered Function
21. Function Live Monitoring • Function Live Monitoring
22. Azure Functions Tools for Visual Studio • Edit, build, and run functions locally. • Publish directly to Azure. • Use attributes to declare function bindings directly in the C# code. • Develop and deploy pre- compiled C# functions. • Included in the Visual Studio 2017 version 15.5
23. Azure Function App • Based on .NET class libraries • Get the full power of IntelliSense, unit testing, and local debugging
24. Visual Studio 2017 Tooling • Create Timer Triggered Function
25. DEMO 2 Azure Functions BlobTrigger AI data processing
26. Function App with Visual Studio
27. Used APIs and libraries • Faces API • Project Oxford SDK. • CoreCompat GDI (our Azure Functions are in .NET Core).
28. BlobTrigger • Triggers function execution when blob is created • Writes the output to a blob
29. Azure Functions hosting • Consumption plan • Instances are dynamically added and removed. • Pay only when your functions are running. • Scale out automatically, even during periods of high load. • App Service plan • Function apps run on dedicated App Service VMs. • Pay for the utilized resources under the App Service plan. • Functions can run longer than the maximum execution time allowed on the Consumption plan (of 10 minutes).
30. Azure Functions scaling • Up to 200 instances. • A single instance may process more than one message or request at a time. • New instances will only be allocated at most once every 10 seconds.
31. Azure Functions billing • Based on per-second resource consumption and executions METRIC PRICE FREE GRANT (PER MONTH) Execution Time* $0.000016/GB-s 400,000 GB-s Total Executions* $0.20 per million executions 1 million executions
32. Micro billing example
33. serverlesscalc.com
34. Usage 1 – as Web API • HTTP Trigger Function == Web API Action * // Endpoint URL: /api/products/{id} public static async Task<HttpResponseMessage> Run( [HttpTrigger(AuthorizationLevel.User, "get", Route = "products/{id}")] HttpRequestMessage req, int id, TraceWriter log) { var msg = new { Id = id, Message = "Hello World" }; return req.CreateResponse(HttpStatusCode.OK, msg); } // Endpoint URL: /products/{id} [Route("products/{id}")] [HttpGet] public async Task<IActionResult> GetProduct(int id) { var msg = new { Id = id, Message = "Hello World" }; return Ok(msg); } * Except that is not!
35. Usage 1 – as Web API • Functions are always static methods. • Functions always receive HttpRequestMessage instance as a parameter. • DI not natively supported! • Benefits in scaling and maybe in pricing due to micro billing.
36. Usage 2 – for scheduled tasks • Only one instance runs at a time. • Executions do not overlap. • Timer history kept in storage.
37. Usage 3 – for backend event-based processing • Email when VM changes. • By using Azure Event Grid. • Process uploaded images. • By using Blob Trigger. • Respond to Service Bus event. • By using Azure Service Bus.
38. Thanks! Any Questions?