Load Balancer for Stretched Cluster Workload Distribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional load balancers fail to intelligently determine which availability zone to use for host servers in stretched clusters, leading to suboptimal workload distribution and performance in active-active and active-passive HA cluster configurations, as they do not consider performance differences across geographical locations.
Innovation Solution
A load balancer executes within a virtual management server, creating and maintaining compatibility rules to ensure active instances of a workload are provisioned on separate hosts across different sites, using a key-value map to rank hosts based on performance data and switch instances to optimize performance and availability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional load balancers are used in stretched clusters, then basic load balancing functionality is provided, but intelligent determination of availability zones for host servers is lost, leading to suboptimal workload distribution
Solution Approach 1:
The load balancer continuously monitors performance data from multiple availability zones and uses this feedback to dynamically determine optimal host server locations for workload placement, enabling intelligent adaptation to changing conditions in stretched clusters
Solution Approach 2:
The load balancer autonomously performs availability zone selection and workload distribution optimization without requiring external intervention, automatically adapting to the stretched cluster environment based on monitored performance metrics
2Ease of manufacture
If active instances are placed on any available host, then provisioning is simplified, but performance optimization and failover capabilities are compromised due to potential placement on the same host or location
Solution Approach 1:
The system applies different placement rules and compatibility requirements to different instances based on their specific performance needs and failover requirements, ensuring each instance is placed on an appropriate host while maintaining provisioning automation
Solution Approach 2:
Compatibility rules act as an intermediary mechanism that automatically enforce separation requirements between active instances without complicating the provisioning process, mediating between simple placement and reliability requirements
3Productivity
If multiple active instances are provisioned without compatibility rules, then resource utilization increases, but the risk of single point of failure increases when instances are on the same host
Solution Approach 1:
The system dynamically adjusts instance placement based on real-time host availability and compatibility rules, allowing flexible resource utilization while automatically preventing single points of failure through enforced separation of active instances
Data Source
AI summary
A balancer can balance workloads in an active-active or active-passive stretched cluster by determining where to provision active instances of virtual machines. The balancer can maintain a set of compatibility rules to govern which hosts, such as sites or servers, are compatible with which instances. The compatibility rules can ensure that the same host does not run both instances, for failover purposes. Additionally, the balancer can create a key-value map that tracks performance of various available hosts. By monitoring host performance and maintaining the compatibility rules, the balancer can re-provision instances on different compatible hosts when performance of a current host suffers.