Link Aggregation and Load Balancing
Link aggregation is a term used to describe various implementations and underlying technologies. In general, link aggregation combines multiple network connections in parallel to increase throughput and provide redundancy. While there are many approaches, this article aims to highlight the differences in terminology.
Link bonding
Link bonding, also known as teaming or bundling depending on the vendor, is generally implemented using two or more links between two logical devices. This could be two servers, two switches, a server to a switch, or various other combinations. Using standards such as Link Aggregation Control Protocol (LACP), the two links are combined into a single logical link, with traffic spread across them evenly. Because link bonding operates at Layer 2, failure detection and isolation can be done quickly, limiting the impact of a link failure. Link bonding is also useful for increasing available throughput between two devices without purchasing more expensive hardware, for example, two 1 Gbps links versus one 10 Gbps link. Refer to Figure 1 below.
Figure 1
Load balancing
Load balancing can also be used to describe link bonding. The term is reserved for Layer 3+ operations. While application load balancers can distribute load across an array of devices for a particular application or purpose, this article focuses on Layer 3. In that context, load balancing is defined as an even distribution of IP traffic across two or more links. This can be done by providing a device multiple equal cost routes to the same destination over equal sized links. Refer to Figure 2 below.
Figure 2
Load sharing
Load sharing is defined as spreading network traffic across two or more equal or unequal links or paths. Load sharing can exist between a 10 Mbps WAN link and a 100 Mbps WAN link. While load sharing often provides the slowest recovery time, dependent on implementation and failure, it is the easiest to implement, most flexible, and still provides levels of redundancy that link bonding and load balancing cannot. As an example, load sharing allows the use of two different ISPs with different link speeds when Network Address Translation (NAT) is implemented. If NAT is not used with other vendor products, this becomes more difficult and may outbound traffic traverse one link while inbound traffic for the same conversation uses another. Refer to Figure 3 below.
Figure 3
Implementation by Cisco Meraki
MS Series
Cisco Meraki MS switches use the open standard LACP to provide Layer 2 link aggregation in the form of link bonding as described above. The MS LACP hashing algorithm uses traffic source and destination IP, MAC, and port to determine which bonded link to use. This provides highly resilient and equal load distribution across two or more links, between two logical devices, with rapid failure detection.
Refer to the Switch Ports for details on how to implement.
When configuring LACP port-channels on uplinks between remote switches or stacks, always configure the remote or downstream side of the port-channel first. Once the configuration has been applied, configure the LACP port-channel on the upstream switch. This insures the downstream switch is not left without connectivity.
*Catalyst 9300-M and MS390 switches will disable a port for 30 seconds if it detects a regular trunk port on the other end of a configured LACP port.
Link Aggregation member ports must operate at the same speed and duplex. It is recommended to use member ports with similar characteristics, including media type where possible, for example, SFP with SFP or copper with copper.
MX WAN appliance
Cisco Meraki WAN appliances use a proprietary algorithm to provide load balancing across two Layer 3 links (if configured). This can be customized to use different ratios and specific rules for outbound traffic. As NAT is used, flows that are part of a particular conversation remain on the link they are placed on.
Refer to MX Load Balancing and Uplink Preferences for details on how to implement.
Configuring link aggregation between MS and Cisco Switches
You may want to set up and configure a bonded link between your Meraki MS switch and a Cisco switch. This is often referred to as link aggregation, link bonding or EtherChannel.
To configure two or more ports, up to eight, as a port aggregate, navigate to Switching > Monitor > Switch ports and select the target ports, then choose "Aggregate". It is recommended that the target ports physically connected to anything during this step.
On your Cisco switch, enable LACP by setting the EtherChannel mode to active or passive depending on the behavior required. For further information, refer to Cisco's documentation on Configuring EtherChannel (this document is for the Catalyst 3000 series).