The client is an American multinational technology conglomerate that develops, manufactures and sells networking hardware, software, telecommunications equipment and other high-technology services and products. To utilize a new Database Technology, the client needed to migrate the Teradata code in their Business Objects Instance to the Snowflake platform.
Owing to the licensing cost, performance issues and an overall mandate to migrate to the cloud, the client was in the middle of a migration of Teradata code to Snowflake platform. To ensure that the business was able to utilize the new database environment, the Teradata specific SQL syntax within the Business Objects environment had to be converted to Snowflake and reintegrated into the Business Objects Environment.
Scope of POC
- Five Web-Intelligence Reports
- 1 Business Objects Universe
- Timeline: One month
- Teradata to Snowflake SemanticLayer Conversion
The Trianz team, consisting of two Snowflake data engineers and one Business Objects Architect, thoroughly analyzed and documented the client’s Business Objects requirements, design and business processes needed to successfully implement the migration. Utilizing Proprietary Code, we extracted the SQL code from the Business Objects Universe, translated it with our custom-designed Evove software solution and then reinserted it into the Business Objects Universe.
SQL Syntax proved to be programmatically successful by converting 90% of the tables, with only 10% manual conversions.
The conversion was tested by running the WebI reports against the new Snowflake environment to ensure that (a) the reports run successfully and (b) the data match. In the live environment, Trianz was able to successfully run all the reports and match the data at the expected rate.
As a result, above 75% of Business Object instances were programmatically converted to Snowflake. Evove was able to automate the conversion of over 95% of derived tables, with only 6 tables converted manually.
Not only was Trianz able to provide initial results on 40% of the reports within the first week, but they were able to complete the migration within a month. Trianz was also able to identify 6 to 7 tables that could be optimized for further Snowflake usage. This process would normally take 2 – 3 months to complete manually, and the solution can be scaled to do 5 to 6 universes within the same time frame, due to the programmatic nature of our application.
|Universe Business Objects||Count|
|No Conversion Needed||~300|
Break-down of Programmatic Errors
|Resolved after compiler fix||Manually fixed items||Underlying database items||No issue items|