EY embraces Azure SQL Managed Instance and unlocks scalability and cost efficiency

Ernst & Young LLP (EY), a globally recognized professional services firm, stands at the forefront of delivering excellence in assurance, advisory, tax, and transaction advisory services. With a commitment to building a better working world, they help clients navigate today's complex business landscape with clarity and confidence.

“Embracing Azure SQL Managed Instance offered us more than just a cloud-based database solution. It helped us unlock a realm of scalability, high availability, and cost-efficiency, providing the foundation for modern, data-driven innovation while seamlessly integrating with the broader Azure ecosystem.”

Naveen Kotakonda, Application Manager, GDS MIS Consulting Reporting, EY

The Global Delivery Services Management Information System (GDS MIS) Consulting Reporting application is used for generating various reports for various business operations teams across GDS. This application was using SQL Server 2012 installed in an on-premises environment. ETL process was written in SSIS packages which were executed on scheduled frequency of daily, weekly, and monthly. The total size of the database was around 1 TB.

Azure SQL Managed Instance – The ideal target

EY has used Microsoft SQL Server 2012 for the GDS MIS application for many years. EY has spent considerable time and resources on training, developing, and integrating the GDS MIS application with other systems. To maximize usage of data, EY wants to spend less time on managing the database and more time realizing the value provided by them. On-premises databases are costly to deploy, manage, and scale, which was posing a significant challenge for business. For example, limited scalability on on-premises databases was affecting the application’s ability to efficiently handle fluctuating workloads, potentially leading to performance bottlenecks during peak periods. Maintenance and patching requirements meant that the team had to factor these into the release cycles, which further constrained their deployment schedule.

"With the small team managing the application, we wanted our team to be more focused on adding value to the business than working on patching, upgrading VMs, and monitoring tasks like backups," explains Naveen Kotakonda, EY's Application Manager, GDS MIS Consulting Reporting.

The core requirements for GDS MIS implied that EY wanted their new database to improve time-to-market, scale on-demand, drive innovation, and provide the ability to pay as you go. EY decided to migrate to Azure SQL Managed Instance which helped EY to reduce time on maintenance, scale as needed, as well as perform Point in Time Restore of database for executing multiple test scenarios. Azure SQL Managed Instance provides the broadest compatibility with SQL Server with all the benefits of a fully managed platform as a service for modernizing existing applications to the cloud.

SQL Server to Azure SQL MI migration approach

The migration of SQL Server 2012 databases to Azure SQL Managed Instance (MI) involved a multi-step process with the use of multiple Microsoft tools.

EY decided to use the database restore as the data migration solution because of its ease of use and performance, as binary backups are optimized for performance. SQL Server Management Studio (SSMS) was used to perform backup of the existing SQL Server Database to Azure Blob Storage. This database was restored to Azure SQL MI using Restore Database from URL process.

EY had multiple SSIS packages running in their current SQL 2012 environment performing crucial data integration, and these were also moved to Azure. By using SQL MI as the target, there was a high degree of compatibility with migrating the existing SSIS packages. Along with this, Azure Data Factory allows SSIS packages as-is on Azure using SSIS-IR on ADF. Visual Studio SQL Server Data Tools (SSDT) was used to perform SQL Server Integration Services (SSIS) packages copy to Azure SQL MI. Azure Data Factory SSIS Integration Runtime (IR) was used to execute SSIS packages in Azure.

After the initial data migration was complete, Azure Data Factory was used to ingest incremental data from the on-premises SQL Server to Azure SQL MI. By running these copy activities in-parallel, they were able to expedite the incremental data movement, and this reduced the overall time by 70 percent.

Overall flow of this process is highlighted in the diagram below:

More about this diagram

Benefits of migrating to Azure SQL Managed Instance

By moving to Azure SQL Managed Instance EY benefited in below areas:

Cost saving: EY realized significant savings cost saving (around 50 percent).
Operational efficiency and savings: Azure SQL MI offerings significantly reduced the amount of time spent by database administrators and developers by eliminating the need to manage infrastructure and platforms and perform maintenance tasks.
Improved business agility: EY can scale seamlessly to meet the changing demands of business with optimal levels of performance, availability, and functionality.

“Earlier business used to ask us, can we provide the database scalability for some duration, which many times was not possible," said Kotakonda. "Now they give us the time for the database to be scaled and we can perform scaling instantly.”

With Azure SQL MI, EY was able to use the built-in process of database backup, built-in high availability, and automatic patching to SQL Engine, which resulted in reducing considerable amount of operational effort.

This simplified process for migrating SQL Server to Azure SQL MI provides a path for further on-premises data tier modernization to Azure SQL in EY.

“With the small team managing the application, we wanted our team to be more focused on adding value to the business than working on patching, upgrading VMs, and monitoring tasks like backups. ”

Naveen Kotakonda, Application Manager, GDS MIS Consulting Reporting, EY