When designing a high availability solution that relies on SAN infrastructure, which three steps should be taken? (Choose three.).

Designing a high availability (HA) SAN solution requires a multi-faceted approach focusing on architecture, validation, and management. The core principle is the elimination of single points of failure (SPOFs) by implementing redundant components and paths (A). The design's effectiveness must be validated by testing failover scenarios under a simulated load to ensure that the recovery time objective (RTO) is met without impacting application performance (C). Finally, comprehensive documentation and topology diagrams are essential for effective implementation, troubleshooting, and lifecycle management, which are critical for maintaining the designed level of availability (E).

B. Practicing manual failovers is a crucial operational readiness and training activity, but it occurs after the design and implementation phases, not as a core step in the design process itself.

D. Joining redundant fabrics with an Inter-Switch Link (ISL) is a design anti-pattern for high availability. This merges two independent fabrics into one, creating a single larger failure domain and violating the principle of fabric isolation.

1. For option A & D: Cisco. (2021). Cisco MDS 9000 Series Best Practices. Section: "High Availability," Subsection: "Redundant Fabrics."

This document states, "The recommended best practice is to have two separate physical fabrics... This design provides high availability because there is no single point of failure at the fabric level... Do not merge the two fabrics with an ISL." This directly supports eliminating SPOFs (A) by keeping fabrics separate and explicitly advises against the action in option (D).

2. For option C: Hewlett Packard Enterprise. (2017). HPE StoreFabric B-series SAN design reference guide. Page 21, Section: "SAN design validation."

This guide emphasizes the validation phase: "The final phase of the design process is to validate the design... Validation testing should include performance testing, failure/recovery testing, and interoperability testing... Failure/recovery testing should be performed to ensure that the SAN will respond as expected in the event of a component failure." This supports the necessity of testing failover under realistic conditions.

3. For option E: Johns Hopkins University, Whiting School of Engineering. (2013). EN.605.446: Storage Area Networks Course Syllabus. Section: "Course Topics."

Reputable academic curricula on SANs consistently include documentation as a key part of the design and implementation process. This syllabus lists "SAN Design and Management" and "Documenting the SAN" as core topics, underscoring that creating detailed documentation and topology is a fundamental step in designing a manageable and robust SAN solution.