Descartes MacroPoint real-time visibility platform migrates to Azure SQL Database to support growth and improve flexibility

In a world that keeps getting smaller and more interconnected, Descartes delivers a wide range of logistics and supply chain solutions that help ensure that companies can efficiently and effectively move products and data across the globe and locally. Descartes provides value to its customers using advanced technology solutions rooted in deep domain expertise. Many of these solutions are network-based, allowing hundreds of thousands of parties to collaborate. As a result, the company was an early adopter of cloud technology.

The Descartes MacroPoint solution provides real-time, location-based truck tracking and predictive truck capacity matching. Transportation brokers, logistics service providers, and shippers use Descartes MacroPoint to track delivery locations of more than 3 million trucking assets nationwide. Customers use the capacity matching features to identify early opportunities to combine loads for more efficient freight moves.

“Our spirit of innovation is one of the things that really differentiates us from our competition,” says Raimond Diederik, Executive Vice President of Information Services at Descartes. “We focus our technology spending on investments that accelerate our ability to innovate and deliver truly game-changing logistics solutions for our customers.”

Recognizing the cloud imperative and choosing the best platform for Descartes MacroPoint

Many of Descartes’ customers have peak seasons, such as the end-of-year holidays, which require a high degree of system resource flexibility. Descartes recognized that public cloud computing is a great alternative to help it meet its varied and growing resource needs while helping it to better focus on the domain-driven capabilities of its solution. Because of its global presence, Descartes wanted to find a partner that could scale to meet its increasing computing needs, and after evaluating its options, Descartes chose Microsoft Azure.

“We had done some early projects in Azure, so we had an understanding of its capabilities,” explains Murray Kristell, Vice President of Descartes MacroPoint Development at Descartes. “For the most part, we’ve been a Microsoft shop—using technologies like .NET and SQL Server—and we felt that Azure had the right combination of services and features for us. Plus, it would be easy to extend our existing Microsoft knowledge to work with Azure.”

Adds Diederik, “From a broader perspective, we have long been a software as a service provider of logistics and supply chain services. Our customers include carriers, transportation and logistics services providers, manufacturers, distributors, and retailers. When we looked at the major public cloud vendors, we saw Microsoft as the company that is highly focused on delivering the breadth of cloud technology services we needed.”

Choosing the right database and service tier for high transaction volumes

Descartes had very stringent public cloud requirements because of the unique nature of the solution. “Descartes MacroPoint is a real-time visibility solution, so by its very nature, it’s constantly processing a lot of time-sensitive GPS location update transactions from hundreds of thousands of sources,” says Diederik. “We live in a true 24/7 business world. Our customers are making critical logistics business decisions and interacting with their customers based on the real-time analysis of their shipment statuses. They expect to have all that data available to them whenever they need it.”

A key element in the planning process for moving Descartes MacroPoint to Azure was selecting the underlying database. MacroPoint ran on Microsoft SQL Server, and historically, a Descartes database administrator manually tweaked and adjusted database parameters and performance. With the move to the public cloud, the company wanted to eliminate the need for manual oversight, so it chose to move Descartes MacroPoint to an Azure SQL Database managed instance. With managed instance, Descartes could gain the system computing and management resources it needed. The company also liked the built-in high availability, automated backups, point-in-time restores, and 99.99 percent uptime service-level agreement (SLA). And it recognized the value in getting all the features of the most recent version of SQL Server along with ongoing enterprise performance enhancements.

Because of the high transaction processing demands of Descartes MacroPoint, the company chose the Business Critical tier available with Azure SQL Database managed instance. “When we tested Business Critical, we saw a night-and-day performance difference from what we were used to,” says Kristell. “We made the easy decision that this was where we wanted to be for ourselves and our customers.”

Enhancing the cloud investment and planning a seamless migration

The move to Azure and SQL Database managed instance inspired the Descartes MacroPoint team to make other important changes to its application ecosystem, such as moving to application orchestration using Azure Service Fabric. “The Microsoft team recommended that we look at orchestration and containerization, so we created a Service Fabric cluster and put some of our services in it, and it worked out really well,” says Kristell. “Then we put our websites into Docker containers and moved them into our environment, where Service Fabric orchestrates them as well.”

Descartes worked with Microsoft Partner Network member Dimensional Strategies Inc. (DSI) on the migration project, and DSI saw managed instance as an excellent option for Descartes. “Because SQL Database managed instance is a single-tenant solution that provides server as a service, it’s a great answer for re-platforming existing applications,” says Stephen Giles, Director of Data Infrastructure at DSI. “If I have existing on-premises systems that I need to lift into Azure, there’s a much lower barrier to entry and a lot less friction when I make the move to managed instance.”

DSI set up database replication for Descartes, and throughout the project, Descartes ran the two systems in parallel, which contributed to a very smooth cutover. “We had less than two hours of downtime for our entire transfer and migration,” says Kristell. “When it was time to go live, all we did was change the connection strings and start everything back up—as simple as that. The project involved a lot of planning, of course, but the performance of SQL Database managed instance and Service Fabric made this one of the best migrations I’ve ever worked on.”

Building on success with new cloud projects, including AI

Descartes MacroPoint went live on Azure near the end of 2018, and everything went off without a hitch. “We saw performance improve at all levels, from the application to our services to our website,” says Diederik. “With Azure, we have the flexibility to quickly scale up resources as needed with just the click of a few buttons.”

And after the busy season ended, the Descartes MacroPoint team found it easy to resize its managed instance in a matter of minutes with no downtime, shifting to a smaller instance that fit its post-peak needs. The success of the Descartes MacroPoint migration project has also spurred new ideas for other public cloud projects at Descartes.

“The Descartes MacroPoint project really accelerated our public cloud plans,” says Diederik. “We’ve got other Azure projects running that are migrations of other solutions out of existing datacenters into a public cloud infrastructure, and we have several more on our radar.”

Descartes is also deploying Azure AI-related solutions, including using Azure Machine Learning service to add intelligence and automation to existing business activities. The company wanted to take advantage of proven AI capability on a common platform, and it’s building a solution with Azure Machine Learning service that helps customers classify product identifiers according to World Customs Organization standards. Descartes MacroPoint is also another beneficiary of AI technology.

“One of the unique capabilities of Descartes MacroPoint is that it can help customers with capacity matching—pairing up shipments that need transporting with trucks that have available capacity,” says Kristell. “We’ve already started using Azure Machine Learning service to help us better match open capacity with loads.”

Building internal skills and looking back at best practices

As Descartes digs deeper into the capabilities of Azure, the company’s IT staff is gaining valuable proficiency with public cloud computing models and resources, which makes it possible to take on larger projects. Descartes has also learned some valuable best practices for other companies to consider when planning their own public cloud migrations.

“We learned the value of upfront planning,” says Diederik. “For example, we have multiple business units and different products, so we prioritized asset classification and tagging, particularly because we expect to remain in our current hybrid situation for quite some time. We also recognize the importance of change management—now that we can deploy infrastructure at will, we need to follow our internal standards and keep everything well monitored.”

Ultimately, Descartes initiated its efforts to move to the public cloud and implement new technologies like AI for one main reason: to improve customer satisfaction. So far, the company is pleased with the impact its new solutions are having.

“Descartes MacroPoint is a great example of how real-time information can change the velocity of logistics and supply chains,” says Robert Derin, Director of Operations at Descartes MacroPoint. “Azure SQL Database provides us with the reliability, scale, and flexibility that allows us to better focus our resources on innovating and improving the Descartes MacroPoint solution.”

Find out more about Descartes on Twitter, Facebook, and LinkedIn.