Infios enhances supply chain management by managing 2,200 database servers on Azure Arc

Infios (formerly Körber Supply Chain Software) provides end-to-end supply chain management solutions to 5,000 global customers. Following a series of acquisitions, Infios now manages more than 20,000 databases.

Administering these databases became increasingly difficult, especially when onboarding new customers. “What sets us apart from our competitors is that our software is highly customizable,” says Tim Radney, Manager of Software as a Service Operations at Infios. “When we bring on a new customer, we go through and fully optimize and tailor our solutions to their business.”

Many of Infios’s customers are deployed on different versions of Microsoft SQL Server, an enterprise database that offers best-in-class security, performance, and availability. Because Infios’s solutions are memory intensive, its customers are prone to database deadlocks, which occur when concurrent transactions permanently block each other because each task has a lock on a resource that the other tasks are trying to lock.

The company’s database administrators troubleshoot deadlocking issues by fine-tuning row-store indexes or turning on “read committed snapshot” isolation in SQL Server, which helps multiple tasks read resources simultaneously and resolves readers-writer locks. If those steps don’t resolve database deadlocking, Infios then escalates the issue to application developers. 

Changing code bases, however, is a time-consuming process that pulls developers away from product development. Infios needed to find a solution that would resolve database performance issues without creating additional work for its developers.

Delivering integrated supply chain management with SQL Server 2025

Infios prioritizes a multicloud approach, deploying customer workloads either on premises or on Microsoft Azure, using Kubernetes for container orchestration. The flexibility supports customer-specific requirements and modernization initiatives, such as deployments for microservices and modular applications, and integration with Azure Arc for governance.

To provide its customers with flexible deployments and support its rapid business growth, Infios wants to modernize its legacy systems and databases to deliver more integrated digital supply chain management solutions. This led its teams to SQL Server 2025, which integrates AI directly into its database engine for more intelligent search capabilities. SQL Server 2025 also features optimized locking, which offers an improved transaction locking mechanism to reduce lock blocking and lock memory consumption for concurrent transactions. 

Additionally, SQL Server 2025 Standard edition offers 256 gigabytes of RAM per instance, which is double the license limit for SQL Server 2022 Standard edition. Because its applications are memory-heavy, Infios’s database administrators often use page data compression to optimize its memory footprint.

By migrating to the latest version of SQL Server, Infios anticipates that it will be able to streamline its third-party tooling. “We can potentially reduce our spend on compute and database licensing costs because we’ll have a higher memory license limit with SQL Server 2025 Standard edition. We’re going to have some flexibility that we didn’t have before,” says Glenn Berry, Services Database Engineer at Infios.

Infios decided to test SQL Server 2025 in private preview and during the Early Access Program. “We joined the private preview not just because some of the older versions are falling out of support,” says Berry. “All the new features in SQL Server 2025 are really going to help us reduce the number of outages we have, improve performance, and make it easier for our teams to manage our databases.”

Achieving zero deadlocks and centralizing database management

Infios’s database administrators combined the optimized locking and optional parameter-plan optimization features of SQL Server 2025 to determine whether they would improve throughput while processing a high volume of transactions. The team then used a custom, automated script to load-test those features and benchmark the performance of SQL Server 2025.

During the load tests, SQL Server 2025 sustained its load with zero deadlocks. Infios anticipates that the feature improvements will free its teams from needing to take extra steps to resolve database performance issues. “Our developers won’t have to make code changes nearly as often,” says Berry. “It’s going to make our workflows a lot easier when we begin migrating customers to SQL Server 2025. I’m really excited about it.”

The company also tested the latest version of the backup compression feature to reduce the size of its backup files by using a new ZSTD compression algorithm. With this feature, Infios can maintain high availability while reducing its CPU usage by 50% during compressed database backups. Additionally, the resiliency of its databases will improve as database administrators tap into the accelerated database recovery feature to perform near-instantaneous rollbacks for long-running transactions.

“The resource governance in SQL Server 2025 that empowers us to prevent runaway queries and excessive disk space consumption is a big improvement for us,” says Radney. “By being part of the Early Access Program, we are able to fully test our products and workloads and thus position ourselves to migrate customers as soon as possible. We are excited to see Microsoft continuing to invest in SQL Server as a mission-critical database.”

The Infios team collaborated closely with Microsoft solutions architects in private preview and during the Early Access Program for SQL Server 2025. During those engagements, the company’s support team learned about Azure Arc, which they deployed to extend their security and governance services to workloads virtually anywhere using an agentless multicloud connector. By digging into Azure Arc’s monitoring and reporting features, the company can apply best practices across all its virtual machines, including ones hosted outside of Azure.

“Our relationship with the Microsoft team has been truly collaborative,” says Radney. “We’ve deployed Azure Arc across 2,200 of our database servers. It has helped us simplify the management of our infrastructure, and we now have better visibility into which versions of SQL Server we’re using. We’re really starting to dig into its capabilities.”

Fueling developer velocity with AI and self-service features

To continue to drive developer productivity, Infios is providing its teams with more AI-powered self-service capabilities. For example, the company is exploring the advantages of deploying Copilot in SQL Server Management Studio 21 (Copilot in SSMS), which can answer questions about its databases and environment and assist developers with writing Transact-SQL using AI. “We have many junior database administrators,” says Berry. “Copilot in SSMS 21 will give them another tool to help them generate high-quality code instead of spending their time looking up code syntax.”

Infios is preparing to migrate its customers to SQL Server 2025 as soon as it becomes generally available to continue to deliver reliable supply chain solutions at scale. “Some of our customers are champing at the bit to get access to SQL Server 2025 and start working with it,” says Radney. “We’re all cheerleading the effort, and we’re telling our product development team to get ready to certify our latest releases on SQL Server 2025.”

Discover more about Infios on LinkedIn and YouTube.