XRC - Extended Remote Copy

Enhanced Definition

XRC (Extended Remote Copy), now commonly referred to as **z/OS Global Mirror**, is an asynchronous remote data replication solution for IBM `DASD` (Direct Access Storage Device) in a z/OS environment. It provides a consistent point-in-time copy of data at a remote site, enabling disaster recovery over long distances with minimal performance impact on the primary system.

Key Characteristics

- Asynchronous Replication: Data updates are written to the primary storage and then asynchronously transmitted to the secondary storage, allowing for long-distance replication without significant performance degradation on the primary application.
- Point-in-Time Consistency: XRC maintains a consistent image of all replicated volumes at the recovery site, ensuring data integrity across multiple volumes and applications at a specific point in time.
- Host-Based Control: The replication process is managed by z/OS software components (e.g., SDM - System Data Mover), which coordinate data movement and ensure consistency across storage subsystems.
- Long-Distance Capability: Designed for distances where synchronous replication (like Metro Mirror) is not feasible due to latency, typically hundreds or thousands of kilometers.
- Recovery Point Objective (RPO): XRC typically has an RPO of seconds to minutes, meaning some data loss might occur during a disaster, but it's generally very low due to the asynchronous nature.
- Performance Impact: While asynchronous, it still requires system resources (CPU, I/O) on the host for managing the replication, though less than synchronous methods.

Use Cases

- Disaster Recovery (DR): The primary use case, providing a remote, consistent copy of critical production data for recovery in the event of a primary site disaster.
- Business Continuity: Facilitates rapid recovery of business operations by having a standby data center with up-to-date data, minimizing downtime after an outage.
- Data Migration: Can be used to migrate large volumes of data between data centers with minimal downtime by establishing replication and then switching over.
- Application Testing: Creating consistent copies of production data at a remote site for testing new applications or upgrades without impacting live systems.

Related Concepts

XRC is a foundational component of IBM's disaster recovery strategy, often integrated into GDPS (Geographically Dispersed Parallel Sysplex) solutions, particularly GDPS Global Mirror. While Metro Mirror provides synchronous, zero-data-loss replication over shorter distances, XRC (z/OS Global Mirror) complements it by offering asynchronous replication for long-distance DR. It relies on DASD storage controllers (like IBM DS8000 series) to manage the physical data copies, with z/OS providing the intelligence for consistency and control.

Best Practices:

Regular DR Testing: Periodically perform full disaster recovery drills to validate the XRC setup, recovery procedures, and RPO/RTO objectives.
Monitor Performance: Continuously monitor SDM performance, DASD I/O, and network latency to ensure the replication backlog remains within acceptable limits and to identify potential bottlenecks.
Proper Sizing and Configuration: Ensure sufficient DASD capacity, SDM resources, and network bandwidth are allocated to handle peak workload replication demands.
Integrate with GDPS: For complex, highly available environments, integrate XRC into a GDPS Global Mirror configuration to automate recovery processes and ensure overall system consistency.
Understand RPO/RTO: Clearly define and communicate the Recovery Point Objective (RPO) and Recovery Time Objective (RTO) for XRC-protected applications, and ensure the configuration meets these targets.