Disaster Recovery Solution in a Hybrid Cloud: OCI Singapore and On-Premise Data Center

This article outlines a robust disaster recovery (DR) solution for a hybrid cloud environment, where the primary site resides in Oracle Cloud Infrastructure (OCI) Singapore region and the disaster recovery site is located in an on-premise data center.

Network Topology:

Both the primary (OCI) and secondary (on-premise) sites utilize a hub-and-spoke network topology. This design provides a centralized point of control and simplifies network management.

Key Features:

Segregation and Security: The architecture segregates Load Balancers, Applications, Databases, and Management components into separate Virtual Cloud Networks (VCNs) and subnets within OCI. This granular approach enables tight security control over communication between different workloads, minimizing the impact of potential breaches.

Hybrid Connectivity: 

FastConnect with redundancy is provisioned in OCI, leveraging Oracle FastConnect partner Megaport. This ensures high-speed, dedicated connectivity between OCI, Azure, on-premise data centers, AWS, and GCP, facilitating seamless data transfer and communication.

Comprehensive Firewall Protection: 

The architecture incorporates both intranet and internet firewalls. These firewalls meticulously inspect and filter all north-south and east-west traffic across different workloads, safeguarding the environment from unauthorized access and malicious activity.

Dedicated Resources: 

Separate VCNs and subnets are dedicated for application servers, load balancers, Exadata Cloud Service (ExaCS), and management components (AD, Jump host). This dedicated resource allocation optimizes performance and simplifies management.

Disaster Recovery Solution:

The core of this DR solution is the database tier, employing Oracle's Active Data Guard technology.

Primary Site (OCI): 

The primary database leverages the power and scalability of Oracle ExaCS, a Platform as a Service (PaaS) offering. This setup includes a 2-node Real Application Cluster (RAC) configuration with Automatic Storage Management (ASM) for high availability and performance.

Disaster Recovery Site (On-Premise): 

To ensure cost-effectiveness, the DR site utilizes a non-ExaCS, single-node database instance with non-ASM storage. This approach strikes a balance between recovery capability and cost optimization.

Application Synchronization:

Application-level data synchronization is achieved using RSYNC, a widely adopted utility for efficient file transfer and synchronization.

Disaster Recovery Activation:

Both planned and unplanned DR scenarios are activated manually. The target Recovery Time Objective (RTO) is 2 hours, and the Recovery Point Objective (RPO) is 15 minutes. This means that the application should be operational within 2 hours of a disaster, with a maximum data loss of 15 minutes.

Uniqueness of the Solution:

This hybrid DR solution, combining OCI and an on-premise data center, particularly the database tier's configuration, presents a unique and potentially groundbreaking approach. It leverages the strengths of both cloud and on-premise environments to achieve a robust and cost-effective disaster recovery strategy.

Conclusion:

This hybrid cloud DR solution provides a comprehensive framework for ensuring business continuity in the event of a disaster. The architecture's emphasis on security, high availability, and cost-effectiveness makes it a compelling choice for organizations seeking a resilient and adaptable disaster recovery strategy.