System performance at Apollo Hospitals improves significantly after migrating to Azure

From legacy systems to a flexible architecture

Apollo’s transformation began with a clear directive: modernize its core hospital systems to support a growing digital footprint and rising clinical demands. Its internal development and IT teams had already built and maintained a custom hospital information system, but the infrastructure behind it was no longer keeping pace. Apollo needed to rethink how it delivered performance, scalability, and flexibility across its network. Security and regulatory compliance were also key drivers of the move. Sridhar Yadla, General Manager at Apollo Hospitals, explains, “The infrastructure was quite fragmented. We had to work harder just to maintain availability, which left very little room for innovation. There was too much firefighting and not enough time to think about what’s next.”

That urgency drove its decision to move from a rigid on-premises environment to Microsoft Azure. One of the first priorities was to replace its Oracle-based backend with Azure Database for PostgreSQL, a choice that reflected Apollo’s commitment to open-source tools and a more future-proof architecture. “The Oracle stack was getting harder to maintain, and scaling it meant significant investment every time,” says Shankar Krishna A. “We saw Azure Database for PostgreSQL as the right foundation for the future. It’s open, cost-effective, and capable of supporting the hospital information system we built in-house.”

Apollo worked with its longtime cloud partner, Quadrant Technologies, to manage the migration. Its development, support, and infrastructure teams coordinated with Quadrant using Azure DevOps to manage deployment, testing, and code delivery across environments. This gave Apollo the structure to keep work moving without downtime or disruption. “Having Azure DevOps in place gave Apollo the control it needed,” says Sreeram Sivakumar, Technical Program Manager at Quadrant Technologies. “There was a lot of coordination across test, staging, and production environments, and the pipelines helped us manage updates without causing delays.”

Apollo’s core application now runs on Azure Virtual Machines, deployed across nine zones. Azure VMs gave Apollo flexibility to lift and shift critical systems while planning for future modernization. The application layer uses Windows-based VMs, while the PostgreSQL database runs as a managed platform as a service (PaaS). Image-based VMs, including Linux on Azure, help standardize configurations, reduce errors, and simplify environment expansion and replication. With fewer infrastructure variables to manage, the hospital has improved application stability and simplified scaling. That foundation also positions Apollo to support more advanced capabilities, including real-time analytics and AI-powered tools like Microsoft 365 Copilot and Microsoft Fabric. “This shift wasn’t just about moving data,” says Sridhar Yadla, General Manager at Apollo Hospitals. “We wanted our team to be ready to support a more scalable and dynamic setup. That meant rethinking how we work and plan for what’s next.”

Handling the volume and complexity of data was one of the most demanding aspects of the project. Apollo’s environment includes more than 80 interfaces connecting the hospital information system with lab equipment, billing, pharmacy, and diagnostics. The team used Azure Databricks to synchronize data between the legacy Oracle system and the new PostgreSQL environment. “Data migration wasn’t just about copying tables,” says Shankar Krishna A. “We had to preserve integrity across thousands of procedures and forms. Azure Databricks helped us keep everything aligned in near real time.” The migration required extensive testing and custom workarounds, including recreating .NET source code from compiled libraries and replacing Oracle’s replication model with a custom solution that minimized lag.

Apollo also strengthened visibility and security. With Azure Application Insights, teams monitor system health continuously and can respond faster to issues. “Database observability was a big gap in our old setup,” explains Shankar Krishna A. “Now with Application Insights, we have much better visibility and can pinpoint issues before they affect our users.”

Operating in a regulated environment, Apollo built its architecture for compliance and secure connectivity. It uses Azure Virtual Network to isolate workloads and Microsoft Defender for Cloud to monitor threats, enforce security policies, and safeguard sensitive data. Sivakumar notes, “The healthcare industry has specific demands for how data moves and where it resides. Using Azure Virtual Network and Microsoft Defender helped ensure we met those from day one.” These tools also give Apollo the scalability to expand securely while maintaining compliance with healthcare data regulations in India.

Adoption was part of the plan from the beginning. Apollo’s internal development team played an active role throughout the migration. That involvement helped reduce friction and set the stage for long-term agility. “Our development team didn’t just watch the migration happen—they were part of it,” says Shankar Krishna A. “They understood the logic behind the move and had the hands-on experience to support it post-migration.”

A platform for the next generation of care

Yadla says the impact on day-to-day work has been significant. “When we were on-prem, tuning and troubleshooting ate up so much of our time,” he says. “Now we’re using Azure Virtual Machines, Azure Database for PostgreSQL, and Application Insights to stay ahead of issues instead of reacting to them. It’s changed how we work.” These tools work together to streamline deployment, monitoring, and performance.

The application layer includes more than 100 applications, mostly legacy .NET with some components in Java. These apps were migrated to Azure and now run on Azure Database for PostgreSQL. Modernization is ongoing, with efforts focused on improving portability, performance, and long-term scalability. The environment spans 200 schemas and packages, with approximately 5,000 tables and 6,500 procedures. In total, the codebase includes millions of lines of code, with some individual procedures exceeding 5,000 lines. “Some of the procedures were over 5,000 lines long,” says Venu Gopal, Principal Architect at Quadrant Technologies. “It was all in Oracle format, and we used our QMigrator tool to convert and optimize it for PostgreSQL.” With everything running in the cloud and integrated into Application Insights, the team can spot issues early and act before they affect performance.

Apollo’s teams are now operating in a more agile environment, with greater control across development, infrastructure, and support. With the cloud foundation firmly in place, the organization is positioned to keep evolving its hospital systems and take on new initiatives faster and with fewer constraints. Apollo is also exploring AI-enabled tools, including a chatbot for appointment booking and potential Copilot integration. The hospital is considering Azure Kubernetes Service (AKS) for future development to support containerized workloads as part of its ongoing modernization efforts. Microsoft offers guidance for organizations on similar modernization paths, including migrating PostgreSQL workloads, automating deployments with Azure Pipelines, and configuring Azure Database for PostgreSQL.

The move to Azure has brought measurable gains across performance and operations. From infrastructure to development to support, Apollo has shifted from maintenance mode to modernization. Its teams are more nimble, and systems are more resilient. With greater stability and faster deployment cycles, staff now spend less time troubleshooting and more time focused on patient care and operational priorities. “We’ve seen nearly 90% of transactions execute within five seconds since the migration, and our uptime has consistently hit 99.95%,” says Yadla. “Opportunities opened up to optimize the operational cost, thanks to decreased overhead and database efficiencies. That stability gives staff the confidence to work without interruption.”

Upgrades and fixes that once took weeks are now completed in days. With Azure DevOps pipelines in place, Apollo can push updates into test and production environments with less risk. “Our deployment timelines have dropped by 40%,” explains Sivakumar. “That means faster access to new features and fewer workarounds for hospital staff.”

Beyond technical improvements, the migration has sparked a shift in mindset. “We’re no longer stuck reacting to problems,” Yadla adds. “Now we’re thinking proactively and looking at how we can evolve.” Shankar agrees that the shift has unlocked new momentum across the organization. He sees Apollo’s foundation as one that’s ready for the future. “We’ve opened the door to future technologies,” he says. “We’re now in a position to build the next generation of hospital systems, not just maintain the current one.”