Link Aggregation Group Load Balancing in Virtual Environments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In virtualized datacenter environments, load balancing across multiple physical or virtual machines often results in a single choke point, reducing network throughput due to the reliance on a virtual load balancer for traffic management within and between host devices sharing link aggregation groups.
Innovation Solution
Configuring link aggregation groups (LAGs) across multiple host devices and a switch to perform load balancing, allowing the switch to distribute traffic between ports and virtual machines, thereby reducing the dependency on a single virtual load balancer and enabling internal load balancing within host devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a virtual load balancer is implemented by a virtual machine running on one of multiple host devices, then load balancing can be provided for a tenant within the datacenter, but the virtual load balancer becomes a choke point for network throughput
Solution Approach 1:
The patent segments the load balancing function from the virtual machine layer to the physical switch layer. Instead of having a single virtual load balancer VM handle all traffic, the physical switch performs load balancing at the network infrastructure level, distributing traffic across multiple host devices before it reaches the virtual machines. This segmentation eliminates the choke point at the virtual load balancer while maintaining load balancing capabilities.
2Ease of operation
If all requests and possibly some or all responses pass through the virtual load balancer, then centralized control is maintained, but network throughput is reduced due to the single point of processing
Solution Approach 1:
The patent introduces the physical switch as an intermediary that performs load balancing before traffic reaches the virtual load balancer. The switch acts as a mediator that distributes incoming traffic across multiple host devices based on load balancing algorithms, while the virtual load balancer continues to provide centralized control for traffic that does require its processing. This intermediary approach maintains centralized control where needed while bypassing bottlenecks.
3Productivity
If multiple virtual machines are assigned to a link aggregation group across multiple host devices, then traffic distribution is improved, but configuration complexity increases
Solution Approach 1:
The patent makes the physical switch universal by enabling it to perform both traditional switching functions and load balancing functions across link aggregation groups spanning multiple host devices. The switch is configured to recognize LAG members from different hosts and apply load balancing algorithms uniformly, eliminating the need for separate load balancer VMs on each host. This multi-functionality simplifies the overall system architecture while maintaining improved traffic distribution.
Data Source
AI summary
Exemplary methods, apparatuses, and systems configure a first set of ports of a host device to be included within a link aggregation group (LAG) with a switch coupled to the first set of one or more ports. A second set of one or more ports of a second host device is also included within the LAG. The configuration of the LAG includes the switch performing load balancing between ports within the LAG. The first host device receives, via the LAG, a packet to be processed by a service implemented by each of one or more virtual machines running on the first host device. The first host device receives the packet as a result of the switch selecting a port within the first and second sets of ports based upon the load balancing between uplinks to the ports within the LAG.