Download to read offline
Exploring Opensource on Microsoft Azure
- 2. Agenda • Introduction toAzure • Exploring OSS • Linux on Azure • Kubernetes • Deploying, Scaling, Securing LAMP Application • Hand on Lab
- 4. Open Source Softwareon Azure From Node.js to Ubuntu, bring your favourite open source software tools and technologies to Azure and open up the possibilities.
- 10. Microsoft & Linux:It’s been a journey
- 11. Open Approach For yourjourney to the cloud Secure Private Transparent Compliant and Certified Freedom of Choice Freedom from Choice Maximize Value Vibrant Local IT Economy X-Platform Open Standards Interoperability Open Source Community Engagement
- 12. Azure VM Regions Availability Managed Disks VMSizes Automation
- 13. CentOS CoreOS DebianOracle Linux Red Hat Enterprise Linux SUSE Linux Enterprise openSUSE Ubuntu
- 14. Create an Ubuntu VM(the 'L' in the LAMP stack) Open port 80 for web traffic Install Apache, MySQL, and PHP Verify installation and configuration LAMP Azure
- 15.
- 16. Allows you toquickly deploy a production ready Kubernetes, DC/OS, or Docker Swarm cluster Azure Container Service
- 18.
- 19. Infrastructure as Code Theprocess of managing and provisioning computing infrastructure and its configuration through machine- processable definition files
- 20.
- 23. BMW enables drivermobility via Azure Service Fabric
- 24. BMW Connected applicationneeded a solution for combining machine learned driver intents, real time telemetry from devices and up-to-date commute conditions such as traffic, monitor the commute needs of each driver
- 25. BMW’s Open MobilityCloud (OMC) Architecture
- 26. Why Azure ServiceFabric • Continually update the service with learned behaviours and commute condition resources • Mobile client limitations like battery and background processing
- 27. Open Mobility CloudAzure Services ?
- 28. Context & ProfileSubsystem
- 29. Role of AzureService Fabric • Scaling millions of drivers made easy • Mobile client limitations like battery and background processing • Rolling upgrades without any downtime • Automatic replication and failover capabilities • Effortlessly roll back to the previous version in case of any error • Company need to focus only the core functionality of their Connected mobile application
- 30. Alaska Airlines improvesservices in the cloud using containers and Azure Service Fabric
- 31. alaskaair.com The systems supportingthe website • enable people to make reservations • look up travel information • manage their mileage plans • get customer support • and more. . . . . . .
- 32. Scale and CapacityIssues • Outgrew their datacentre. • Loss of services caused by • Networking issues • Hardware issues • Virtual machine issues • Performance impacts • Governed usage to maintain uptime requirements
- 33. From on-premises toAzure • Azure App Service for hosting their ported APIs, • Azure API Management for secured access to those hosted APIS • Azure Cache for session management.
- 34. Next Challenge • Designflaw prevented the system from recovering from failures on top of the inconsistent performance of third-party APIs and downstream dependencies • Need of horizontal scale of the whole system beyond the App Service scaling limit
- 35. Solution of Alaska •Uphold their use of VSTS for CI/CD. • Work with limited to no refactoring--there wasn't time. • Run locally, on-premises, and in the cloud. • Support a variety of libraries and technology versions.
- 36.
- 37. Shopping Cart APIarchitecture
- 38. Azure Service Fabric+ Windows Containers • Improved scale and density they needed for their crucial service with greater control • Self-correcting when a node or the services on the node fails • API Deploy time and time for consumption reduced • 20 minutes in Azure App Service to 10 seconds in service fabric • In-place upgrades with automatic rollback • Savings at scale
- 39. “We love thepristine and predictable environments provided by Windows Server Containers. Service Fabric brings us a safe and consistent deployment strategy that maximizes uptime. The marriage of the two in Microsoft Azure gives us unlimited potential” Ralph Feltis: Software Engineer Alaska Airlines
Editor's Notes
- #23 Middle one – design ideas
- #26 Enabling Driver Mobility via Service Fabric In order to proactively act on behalf of a user, the BMW Connected application needed a solution for combining machine learned driver intents, real time telemetry from devices and up-to-date commute conditions such as traffic. In addition, this service was built to scale up to millions of drivers, and asynchronously monitor the commute needs of each driver. Furthermore, BMW wanted to continually update the service with learned behaviors and commute condition resources. Because of these reasons as well as mobile client limitations (battery and background processing) they found that the Service Fabric Actor model was the right solution for their service.
- #28 Enabling Driver Mobility via Service Fabric In order to proactively act on behalf of a user, the BMW Connected application needed a solution for combining machine learned driver intents, real time telemetry from devices and up-to-date commute conditions such as traffic. In addition, this service was built to scale up to millions of drivers, and asynchronously monitor the commute needs of each driver. Furthermore, BMW wanted to continually update the service with learned behaviors and commute condition resources. Because of these reasons as well as mobile client limitations (battery and background processing) they found that the Service Fabric Actor model was the right solution for their service.
- #38 The Shopping Cart API wasn't built on the .NET Core, so it wasn't host-independent. The API required Internet Information Services (IIS), which isn't supported as an application host, since Service Fabric expects workloads to be self-hosted. The solution to this challenge involved yet another technology introduced to the team at the hackathon--Windows Server Containers.