When designing a redundant IP storage network for data movement or access of remote data, which two must be considered? (Choose two)

When designing a redundant IP storage network for remote data, the primary goals are ensuring high availability and consistent performance. Option C directly addresses these by specifying low-latency, high-quality circuits with diverse physical routes. Diverse routing is the cornerstone of redundancy, as it protects against a single path failure (e.g., a fiber cut). Low latency is critical for storage protocols, especially for synchronous replication over distance. Option D is a foundational engineering principle; a successful and resilient design is impossible without a thorough understanding of the capabilities, limitations, and management of all network components, which ensures they can support the required redundancy and performance characteristics.

A. Obtaining the fastest available circuit focuses solely on bandwidth, not redundancy. A single, high-speed link remains a single point of failure, which is contrary to the goal of a redundant design.

B. Jumbo frames and TCP window sizing are performance tuning optimizations. While beneficial, they are not fundamental architectural considerations for designing network redundancy and are implemented after the core design is established.

1. Cisco Systems, "Data Center Interconnect Solution Overview," White Paper, 2016.

Reference: Page 3, Section "DCI Requirements." The document states that a key requirement for Data Center Interconnect (DCI) solutions, which carry storage traffic, is "Resiliency." It elaborates: "The DCI solution must be able to provide resiliency through multiple, diverse paths between data centers and fast convergence in the event of a path failure." This directly supports option C.

2. IBM Redbooks, "Introduction to Storage Area Networks," SG24-5470-08, Ninth Edition, March 2017.

Reference: Page 178, Section 5.3, "FCIP networking considerations." When discussing extending storage networks over IP, the guide emphasizes, "The quality and performance of the IP network is critical to the performance of the FCIP link. The two most important factors are the latency of the IP network and the percentage of packet loss." This supports the "low latency and good quality" aspect of option C.

3. Prasad, V., & Kumar, V. (2014). "Data Center Network Architectures." In Network Innovation through OpenFlow and SDN: Principles and Design (pp. 249-279). CRC Press.

Reference: Chapter 10, Section 10.2, "Requirements of Data Center Networks." The text highlights that data center networks must provide high availability and reliability. It discusses multipathing as a key technique to achieve this, stating, "To avoid single points of failure, it is important to have redundant paths between any two servers." This academic source validates the principle of diverse routes in option C.

4. Cisco Systems, "Campus Network for High Availability Design Guide," August 2019.

Reference: Chapter 1, "Introduction," Section "Designing for High Availability." The guide states, "A successful high availability design requires a holistic approach... It is important to understand the capabilities of the devices, the features that are available, and how to use them to prevent and recover from failures." This principle is universally applicable and directly supports option D.