Design a bidirectional data sync dashboard

Design a **bidirectional data synchronization platform** and an internal **dashboard**. ### Scenario Your company integrates with multiple cloud providers (e.g., AWS/Azure/GCP-like). Each provider can have planned maintenance windows and unplanned incidents/outages. After a provider experiences or announces maintenance/outage, the status should be reflected on **your own dashboard**. Additionally, certain configuration/status data must be kept in sync **both directions** between: - **Provider side** (provider APIs/events) - **Your side** (your internal system-of-record) ### Requirements 1. **Bidirectional sync** of selected objects (you define what objects make sense, e.g., “account config”, “connection status”, “maintenance events”). 2. **Ingestion of provider events** about maintenance/outage and updates to your dashboard. 3. Handle **expected provider maintenance/outage** (planned) and **unexpected outages**. 4. Should work across multiple providers/tenants. 5. Reliability: retries, idempotency, ordering/duplication issues. 6. Define how you detect changes on both sides (polling vs webhooks vs change logs). 7. Show how the dashboard queries/gets the latest state. ### Deliverables - High-level architecture (components and responsibilities) - Data model for synced entities and outage/maintenance events - Sync protocol: conflict resolution strategy and consistency guarantees - Failure handling (provider API downtime, partial failures, replay) - Key metrics/alerts and operational considerations - Brief note on security (auth, secrets, tenant isolation)