.Database Engagement in AWS – Schema Optimization, Data Capture, Roles Mapping
Scope
• 2019 MS SQL servers are EC2 instances on AWS, (no RDS).
• Setup change data capture for FIMMAS database.
1. The FIMMAS databases use unique indexes and have no keys, making this a very manual setup.
2. About 1100 tables.
• Schema optimization, to towards segmentation of data
1. Create schemas.
1. Analysis of separation
2. Create schemas
2. Transition CRUD views
1. Analysis of need views and what schema they should go under
2. Create views.
• Roles mapping to schemas to provide better granularity on security.
1. Creation of roles for each schema
1. CRUD
2. View Def
3. Execute
2. Identify common grouping of roles.
• Security create AD groups map to roles.
• Security clean-up and get added to attestation system.
1. Walking SR/CRs through IAM.
2. Analysis for linking to attestation/AD groups
• Structure complete partitioning.
1. Define common indicators for horizontal partitioning
2. Create jobs to trigger movement.
• Ensure optimization routines are scheduled and running (scripts).
• Enable database email functions.
• Cleanup code structure so script extracts are readable. Issues and need -
1. Get well defined code structure.
2. Low level tech goes through every view and Stored proc and makes adjustments or confirms.
3. There are about 420 scripts current, it might double when Wynsure views get added.
• Design process to load data from UAT to lower environment. Issues and need -
1. Keep lower environment security in place.
2. Have processes but the build out takes time, and confirmation of timing when it is ran.
3. Documenting the process so someone who comes after can run.
• Setup auto rerouting of connection to and between Prod ó Read Only Prod.
1. Current setup has non-clustered read only replicas.