08 What is Network Redundancy in Business?
A Comprehensive Explanation of Its Benefits and Types

The term 'redundant' might sound negative in everyday speech, implying something is excessively long and filled with unnecessary elements. However, in the world of the internet, it takes on a different, positive meaning.

This article explains what redundancy in networking is, the different methods of achieving it, and its importance in a way that's easy for beginners to understand.

What is Network Redundancy?

Network redundancy, or redundant configuration, in networking, refers to preparing backups in case of failures. For communication lines, it means setting up multiple routes from point A to B; for communication equipment, it involves having multiple pieces of the same equipment ready. Redundancy ensures that network outages have minimal impact on service continuity and allows networks to maintain stable service provision.

Benefits of Network Redundancy

Redundancy functions as a critical component of BCP^*1. Backup equipment and lines are often located geographically far from the primary locations, especially when made with natural disasters in mind. This ensures that the business can continue using the remotely located backup facilities even if a major facility is damaged in a natural disaster.

*1 BCP (Business Continuity Plan): A plan for continuing business operations.

Types of Network Redundancy

There are two main types of network redundancy: hot standby and cold standby. These methods aim to enhance network availability and recovery from failures. Below, we explain the characteristics of each type.

Hot Standby

Hot standby involves keeping backup equipment powered on even when not in use. This minimizes network downtime and enhances service continuity. However, operational costs can be higher as backup equipment also runs simultaneously.

Cold Standby

In cold standby, backup equipment is kept powered off while on standby. In case of a failure, equipment is manually switched to the backup. This method reduces costs as equipment doesn't need to be connected to the network at all times. However, as a drawback, it requires manual effort and takes time to prepare.

Layer-based redundancy

Layer-based redundancy involves implementing redundancy for each layer of network functions and system configurations. This ensures that if a failure occurs in a particular layer, other layers continue functioning, improving overall network availability and speeding up failure recovery.

Typically, redundancy is implemented in layers like the physical layer and data link layer. We will explain specific methods for each layer.

【Layer1】Physical Layer

Physical layer redundancy involves duplicating network equipment, servers, cables, and other physical systems. Power redundancy is also used, which involves equipping servers with multiple power supplies.

【Layer2】Data Link Layer

Data link layer redundancy includes methods like teaming, link aggregation, and Spanning Tree Protocol (STP).

• Teaming: A method to virtually unify multiple NICs^*2 inserted in one device.
• Link Aggregation: Utilizing multiple communication paths so communication can continue even if one path has issues.
• Spanning Tree Protocol (STP): A mechanism to prevent data from looping infinitely in a loop-structured network.

*2 NIC (Network Interface Card): An expansion device to connect computers and other devices to a communication network.

【Layer3】Network Layer

Network layer redundancy includes methods like VRRP and ECMP.

• VRRP: Combines multiple routers on a network into a single virtual entity.
• ECMP: Distributes the load across all communication paths when there are multiple equal-cost paths.

【Layer4】Transport Layer

Transport layer redundancy includes HA and DSR.

• HA: Combines two firewalls into one, designating one as the operating device and the other as a standby.
• DSR: Returns traffic directly from the server to the client without going through a load balancer.

Conclusion

Implementing redundancy enhances the reliability of a network. It's important to choose the right technology, set it up correctly, and operate it effectively for each layer.