Data Flow & Component Breakdown
The following matrix details where specific data types are stored and the security controls applied to each persistence layer.
Component | Technology | Hosting Provider | Data Classification | Security Posture |
|---|---|---|---|---|
Primary Data Store | MongoDB Atlas (Enterprise) | AWS (VPC Peered) | Golden Records, PII, User Profiles, Transaction Logs, Raw Metadata. | Network: PrivateLink / VPC Peering. Encryption: AES-256 (At Rest), TLS 1.3 (In Transit). Access: IP Whitelisting & IAM Role Integration. |
Vector Engine | Pinecone (Enterprise) | AWS (Managed) | Vector Embeddings (Float Arrays), Anonymized Semantic Indexes. | Network: Private Endpoints. Isolation: Single-tenant pod isolation (Enterprise tier). Compliance: SOC 2 Type II. |
Graph Relationship Store | AWS Neptune | AWS (Private Subnet) | Entity Relationships, Corporate Hierarchies, Identity Graphs. | Network: Deployed strictly within private subnets (No Public Internet Access). Auth: IAM Authentication. Encryption: AWS KMS Managed Keys. |
Performance Cache | Redis Cache (Redis) | AWS (Private Subnet) | Session Data, Transient Query Caching, API Rate Limiting keys. | Network: Private Subnet. Lifecycle: Ephemeral (TTL based). Encryption: Encryption in transit and at rest enabled. |