Multi-tenant Feedback Controller Weights
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In multi-tenant cloud computing systems, existing technologies face challenges in dynamically adjusting weights for selecting requests from tenant queues to ensure optimal resource allocation and performance isolation, particularly in managing response time and throughput violations across tenants.
Innovation Solution
The method involves determining weights for each tenant based on measured response times and throughput, using initialized weights, modified PI control, and volunteer weight increases, and normalizing these weights to ensure fair resource distribution and compliance with service level agreements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If a single application instance is shared among multiple tenants to reduce costs, then resource utilization and cost efficiency are improved, but performance isolation and response time guarantees for individual tenants deteriorate
Solution Approach 1:
The patent implements differential quality of service by assigning different weights to different tenants based on their specific performance requirements and SLA agreements. The weight-based scheduling mechanism allows each tenant to receive customized resource allocation, ensuring that tenants with stricter performance requirements get prioritized processing while still sharing the same application instance, thus maintaining both cost efficiency and performance isolation.
2Ease of operation
If static weights are used for tenant queue selection, then system simplicity and ease of operation are improved, but adaptability to changing performance requirements and SLA compliance deteriorate
Solution Approach 1:
The patent implements dynamic weight adjustment where tenant weights are not fixed but are continuously adapted based on measured performance metrics and SLA compliance status. The system monitors response times, throughput, and error rates, and automatically adjusts weights to maintain SLA compliance while adapting to changing workloads and performance requirements, thus achieving both operational simplicity and dynamic adaptability.
Solution Approach 2:
The patent employs a feedback mechanism where performance metrics (response time, throughput, error rates) are continuously measured and fed back to the weight adjustment algorithm. This closed-loop control enables the system to automatically respond to performance deviations and adjust tenant weights accordingly, ensuring SLA compliance without manual intervention while maintaining system adaptability to changing conditions.
3Reliability
If aggressive weight adjustment is applied to enforce SLA compliance, then response time guarantee and reliability are improved, but system throughput and productivity may deteriorate
Solution Approach 1:
The patent applies partial weight adjustment by only modifying weights for tenants that are violating their SLAs, rather than aggressively adjusting all tenant weights. This selective approach allows the system to maintain SLA compliance for affected tenants while minimizing disruption to overall system throughput and maintaining productivity for compliant tenants, thus resolving the contradiction between reliability and productivity.
Data Source
AI summary
Methods, systems, and computer-readable storage media for determining weights for selecting requests from a plurality of tenant queues in a multi-tenant system receiving measured response time and measured throughput for each tenant in a set of tenants being considered in a current period, for each tenant in the set of tenants, determining a weight based on respective measured response times and respective measured throughput, the weight being determined based on one of a previous weight, an initialized weight, a modified proportional and integral (PI) control, and a volunteer weight increase, providing a set of weights that includes weight for each tenant in the set of tenants, and transmitting the set of weights to an access mechanism, the access mechanism selecting tenant requests for processing by a shared resource.