Database Refactoring
Restructure schemas, normalize or carefully denormalize, and sequence migrations so applications keep reading and writing while the model changes.
Flagship engagement →Web Maple Core
We work with engineering and data teams in Taiwan and abroad to untangle schemas, shorten slow queries, and plan migrations without overnight firefighting as the default plan.
What we take on
Each engagement starts with the database you already run: tables, indexes, stored procedures, and the queries your application actually issues under load.
Restructure schemas, normalize or carefully denormalize, and sequence migrations so applications keep reading and writing while the model changes.
Flagship engagement →Trace slow statements, examine plans and indexes, and leave you with ranked fixes your team can apply in sprint order.
Details →A time-boxed read of naming, constraints, keys, and growth patterns — useful before a major release or when hiring a new DBA.
Details →How work begins
We ask for anonymized query logs, schema dumps, or staging access — never production credentials on the first call. You receive a proposed timeline, risks, and what we will not touch.
From recent engagements
“They rewrote our nightly reporting join so it finished before the warehouse window closed. The first week still needed our ops person on standby, which they warned us about.”— Mei-Ling Chen, Engineering Lead, Taipei retail analytics team
“We had twelve years of ad-hoc columns. Web Maple Core mapped which ones still mattered and which could be archived — the migration checklist was clearer than our internal wiki.”— Marcus Ong, Product Operations, Kaohsiung logistics firm
Tell us which engine you run, what hurts under load, and when you need a change window. We reply within two business days.