Migration from Teradata to AWS Cloud
Globally, organizations are working towards modernizing their IT systems to harness the benefits of improved performance and on-demand scalability. For a global motor-vehicle manufacturer working with on-premise Teradata database servers, this was a step that could make all the difference.
With Birlasoft as their strategic IT partner, they successfully migrated their on-premises Teradata database to AWS cloud and experienced significant savings in licensing costs and made rapid strides in efficiency and innovation.
The Challenges
Sunset of Teradata Database Servers
With Teradata being archaic, and the problem further exacerbated with high licensing costs and outdated capability for analytics, the client desired to migrate to an efficient cloud-based environment. However, the challenge here was migrating from a system that the client had been reliant on for a long time.
Dealing with Increasing Data Volumes
With an on-premise database, the client grappled with the challenge of scalability to deal with data volumes increasing year on year. There was also the additional disadvantage of dealing with the costs of running the traditional datacenter and maintaining the physical infrastructure needed for the same.
Tackling Performance and Down Time
The client faced challenges in improving performance and desired to reduce down-time. To ensure business continuity with resilience, a scalable cloud infrastructure was the way to go to accelerate performance without compromising on the reliability and availability.
The Solution
Making the Switch to AWS Cloud
Birlasoft was instrumental in the migration to AWS cloud, building of the solution architecture, converting Teradata objects to Postgres objects, and migrating the current as well as historical data of up to 10 years.
Birlasoft did an exhaustive exercise to migrate a high performing and high available transactional system on to AWS cloud. As part of this exercise, database usage, IOPS and workload characteristics on the source Teradata database were analyzed to suggest the right database, DB instance type and storage needs in order to suggest the right approach and strategy for database migration to RDS Postgres Instance.
This involved the migration of Teradata DB objects including tables, stored procedures, functions, views, and underlying data over the client cloud hosted on AWS platform on Postgres Aurora Global DB using fully automated Migrate tool, SSIS Packages and stored procedures over a dedicated AWS direct connect network. Birlasoft automated the entire Data Load process for the BAU activity using a Migrate Orchestrator, a scheduler and python code.
The AWS services used were: Amazon RDS for PostgreSQL, AWS direct connect, Route 53, AWS S3 for backups, and EC2.
The Impact
By making the shift to AWS Cloud, the client harnessed the host of benefits offered by cloud-based environments. There was a significant improvement in performance and savings with reduced licensing costs and FTE savings. The down time risk was minimized, and business continuity was maintained.
- Apart from the initial investment in cloud, over a period, the client stood to experience cost savings up to 30% due to the cost benefits associated with IaaS versus on-premises.
- ROI of the overall project was 13.44%
- Up to 50% enhancement in business flow completion was observed.
- There was a zero down time risk due to multiple read replicas and multi region replication to help recover database whenever needed.
- Enhanced ability to deal with increasing demand with scalable data storage and archival.
- The client was able to cut back on licensing costs and additionally experienced resource cost savings up to one FTE per month.
