Strategically led the datamodel changes inside the Salesforce service. Partnered with DBAs, operations teams, QA, and development resources to design and rollout methods for quickly and efficiently deploying datamodel changes. Enabled the company to successfully deliver new major releases in a predictable cadence with minimal downtime for customers.
Championed a yearlong project that re-implemented the database schemas using Oracle’s hash partitioning system to overcome some of the limitations of the Oracle 9i RAC product. Realized huge success in the partitioning method and standardized the method as the layout for all production Oracle databases.
Designed and assisted in coding a tool that enabled database changes to be deployed and controlled in a repeatable fashion. Extended the tool’s use in a simple way to support the deployment of changes to Postgres as well as Oracle databases.
Migrated toward an approach that made database triggers version aware. Slashed downtime associated with each release from several hours to under one hour.
Undertook significant changes that enabled the execution of data and datamodel changes to seamlessly occur in a hybrid Oracle-Postgres architecture.
Actively worked on a large project that created a single set of libraries for Developers to deploy DML in a database agnostic fashion. Took DML and calculated primary key ranges. Committed the DML statement in discrete chunks. Leveraged the advantage of certain Oracle-specific features that included dms_parallel_execute, and Rowid ranges.
Leveraged the editions based redefinition process that eliminated all downtime for the rollout of each new version of the Salesforce service. Enabled Salesforce to sell to new markets that required near 100% system uptime.
Created libraries to run complicated DML statements in both database architectures. Required minimal Developer efforts for the efficient migration of large production databases.
