Converged Analytics Deployment

Service Overview

The EDB Converged Analytics Deployment engagement delivers an installed, configured, and validated Converged Analytics environment on EDB Postgres AI. This engagement deploys the Analytics Engine, Lakehouse Connector, Iceberg REST Catalog, and data source connections — enabling your organization to run analytical queries directly against operational data without complex ETL pipelines.

This is a deployment-focused engagement: EDB installs and validates the technical infrastructure. When complete, your environment is ready for analytical data products and business intelligence workloads. For guided use case implementation, see the Converged Analytics Enablement engagement.

Scope of Service

Onboard

• Kickoff to confirm scope, validate infrastructure access, and review data source inventory
• Review existing database estate and identify source systems in scope
• Review data governance requirements and access control policies

Installation

• Deploy Postgres Analytics Accelerator (PGAA v1.9) 
• Configure PGFS storage locations for each object store in scope
• Set up Iceberg REST Catalog integration for each configured storage location — enables schema discovery, table metadata management, and time travel queries
• For Medium+: deploy Conversational Analytics interface (NL-to-SQL endpoint)

Configuration

• Configure data source connections with appropriate credentials and read permissions
• Set governance policies per data source (analytics enable/disable, column masking, PII exclusions)
• Configure Iceberg table replication strategies and refresh schedules
• Establish role-based access controls for data consumers

Validation

• Validate end-to-end data flow from each source to Iceberg tables
• Confirm data governance policies are enforced correctly (masking, exclusions)
• Run analytical query benchmarks against deployed Iceberg tables
• Test PGFS Connector connectivity and replication
• Demonstrate Analytics Engine query performance with customer data
• Review and document Standard Operating Procedures

Project Closure

• Formal sign-off review against agreed scope
• Deliver configuration documentation and operational handoff

Deliverables

• Fully installed and operational Converged Analytics environment
• Configured PGFS Connectors for all agreed data sources
• Iceberg REST Catalog with validated data replication
• Documented data governance policies and access controls
• Analytics Engine configuration documentation and operational SOPs

Roles and Responsibilities

EDB Project Manager: Responsible for engagement planning, prerequisite coordination, schedule management, and closeout.

EDB Solution Architect: Designs the analytics architecture, reviews data governance requirements, and validates configuration against EDB best practices.

EDB Senior Consultant: Technical lead for all installation, configuration, validation, and documentation activities.

Customer Team: Provides system access and participates in validation. Key roles:

• Data Owner / DBA for each source system
• Infrastructure / Platform Engineer
• Security / Compliance Officer (for governance configuration)
• Network Administrator

Assumptions

• A kickoff call will confirm scope and schedule prior to work beginning.
• This engagement is delivered remotely unless otherwise agreed.
• EDB Hybrid Manager is deployed and operational prior to engagement start.
• EDB Postgres AI licenses are in place for all components being deployed.
• Object storage (S3-compatible, Azure Blob Storage, or Google Cloud Storage) is provisioned and accessible prior to engagement start.
• Data governance requirements and PII column inventory are defined by the customer prior to configuration.
• PGAA reads data from object storage in open formats (Iceberg, Delta Lake, Parquet). Source data must already reside in, or be replicable to, object storage before analytical queries can run.
• Delta Lake integration provides read access only; write operations to Delta Lake tables are out of scope.
• Live production traffic on analytical workloads is out of scope during deployment.
• Customer will provide timely feedback on deliverables. Items without commentary within 5 business days are deemed accepted.
• Customer will not provide Personal Data as defined by applicable law.

Prerequisites

• Postgres databases are identified and accessible. 
• Object storage is provisioned and credentials are available.
• Customer network permits connectivity between Analytics Engine and all target data sources.
• Data source owners have been identified and are available during the engagement.
• Column-level governance requirements (mask/exclude) are documented before configuration begins.
• Storage type must be: S3-compatible (including MinIO), Azure Blob Storage, or Google Cloud Storage

Validation Test Cases