AdTechCo manages escalating customer demands
efficiently with an Auto-scalable DevOps solution

About the Client

The client is a New York based advertising technology company who created the largest in-app mobile network. They recently built a real-time programmatic advertising platform that connects Western and Chinese advertisers seamlessly into the growing advertising market of China.

Our Results

The Continuous Integration solution helped the client deploy code on the virtual machine scale set in less than 30 minutes. The machines can now scale automatically whenever the workload goes beyond or below a certain threshold limit, allowing the customer to meet huge demands reducing the need to expend towards resource management.

Read More

Need an auto-scalable DevOps Solution for your business?

Talk to our DevOps Specialist

View Other Case Studies

Business Need

The Client is building an Ad-Exchange Java based application and wanted to deploy it on an auto-scaling infrastructure that can handle millions of users from China. They wanted to host their application on a tomcat server in Azure environment, which could scale out to multiple VMs depending on high workloads.

As the application development is still in progress, they wanted their latest application build to reflect in the scaled architecture. Any change in code should replicate across all VMs. They also needed an automation script to spin-up similar environments on-demand.

Technology Solution

Saviant’s DevOps team engaged with the client to assess their project requirements and devised a strategy to solve their problem.

The DevOps process included version control, Continuous Integration, Infrastructure automation and Configuration management services. Tools such as Packer, VSTS (Visual Studio Team Services) and Ansible were leveraged to ensure that continuous deployment happens on the VM scale set.

As a part of setting-up the environment, the continuous integration consulting team at Saviant linked the application to GitHub and VSTS Source Control and connected Azure to a VSTS account. They leveraged Packer to create the VHD image, baked with the application code. Scripts were written to use the VHD created by Packer to deploy on Azure.

A VSTS build is triggered each time a code is merged into the master branch of the GitHub repository. The VSTS build just publishes the files needed for the release. Successful builds then trigger a release. The first step of the release packages the latest application files into an Azure VHD image using packer, this VM image (vhd) is stored in the configured storage account. Ansible was used to configure the image. The team used the Azure CLI group deployment command to push the new VM image (with latest application code baked in) to the VM Scale set. At the end of each successful release, the VM scale set had the latest application code.

By adopting Continuous Integration, Infrastructure automation and Configuration Management, the client could successfully deploy its latest app to the Virtual Machine Scale Set.

DevOps Continuous integration solution architecture

Utility giant cuts time-to-market by 50% with DevOps

An independent AMR provider for Utilities adopts DevOps that helped in increasing release velocity by 60%, reduced deployment failures and faster development cycle. The client now has an integrated system that provides simple version control, robust build & release management & easy-to-use application life cycle management.

Customer Success Story

Continuous Integration Vs Continuous Delivery Vs Continuous Deployment

Continuity is the cornerstone of every DevOps strategy. DevOps calls for process continuity in the form of Continuous Integration, Continuous Delivery & Continuous Deployment. Let’s take a closer look at each one of these.

Read the Blog

Want to build custom Azure DevOps solution?

Talk to our DevOps Specialist