Inline Load Balancer in Source Compute Node Egress
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current load balancer deployments in datacenters face issues as chokepoints due to network traffic bottlenecks, require manual configuration, and fail to seamlessly scale with increasing or decreasing data traffic, limiting their ability to efficiently distribute load across compute nodes.
Innovation Solution
Implementing an inline load-balancing method where a load balancer is deployed in the source compute node's egress datapath to dynamically balance data traffic across multiple destination compute node groups by changing destination addresses and using a distributed load balancing architecture that adjusts load balancing criteria based on real-time statistics, allowing seamless scaling and efficient traffic distribution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If load balancers are deployed at the edge of the network as standalone machines or service VMs, then traffic load can be distributed to multiple computing resources, but the load balancers become network traffic bottlenecks and chokepoints that limit scalability
Solution Approach 1:
The patent extracts the load balancing function from traditional edge-deployed load balancers and relocates it to the egress datapath of source compute nodes. This removes the centralised bottleneck by distributing the load balancing functionality across multiple source compute nodes, eliminating the chokepoint problem while maintaining traffic distribution capability
Solution Approach 2:
The patent segments the load balancing function into distributed components embedded within source compute nodes rather than using a single centralized load balancer. Each source compute node independently performs load balancing for its own egress traffic, dividing the overall load balancing task across multiple locations in the network
2Productivity
If traditional load balancer deployments are used, then traffic can be balanced across computing resources, but manual configuration is required and the system cannot seamlessly scale with increasing or decreasing data traffic
Solution Approach 1:
The patent implements dynamic load balancing by embedding the load balancer in the source compute node's egress datapath, allowing it to automatically adapt to changing traffic conditions. The system dynamically adjusts load distribution based on real-time statistics and conditions without requiring manual reconfiguration, enabling seamless scaling as traffic increases or decreases
Solution Approach 2:
The patent incorporates feedback mechanisms where the distributed load balancer monitors traffic conditions and computing resource states in real-time, then automatically adjusts load distribution decisions. This feedback loop enables the system to seamlessly adapt to changing traffic patterns and scale dynamically without manual intervention
3Adaptability or versatility
If load balancers are deployed in the source compute node's egress datapath, then manual configuration is eliminated and seamless scaling is enabled, but the system must dynamically determine and change destination addresses for each data message
Solution Approach 1:
The patent applies preliminary action by pre-establishing the load balancing logic and destination group mappings within the source compute node before traffic arrives. The system prepares the egress datapath with the necessary routing information and load balancing rules in advance, enabling rapid real-time address determination without complex runtime calculations
Data Source
AI summary
Some embodiments provide a novel method for load balancing data messages that are sent by a source compute node (SCN) to one or more different groups of destination compute nodes (DCNs). In some embodiments, the method deploys a load balancer in the source compute node's egress datapath. This load balancer receives each data message sent from the source compute node, and determines whether the data message is addressed to one of the DCN groups for which the load balancer spreads the data traffic to balance the load across (e.g., data traffic directed to) the DCNs in the group. When the received data message is not addressed to one of the load balanced DCN groups, the load balancer forwards the received data message to its addressed destination. On the other hand, when the received data message is addressed to one of load balancer's DCN groups, the load balancer identifies a DCN in the addressed DCN group that should receive the data message, and directs the data message to the identified DCN. To direct the data message to the identified DCN, the load balancer in some embodiments changes the destination address (e.g., the destination IP address, destination port, destination MAC address, etc.) in the data message from the address of the identified DCN group to the address (e.g., the destination IP address) of the identified DCN.