Workload Manager Evaluator for QoS and Resource Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a difficulty in evaluating the performance of workload managers, specifically in determining optimal control parameters for schedulers, as no effective metric is available to assess the performance under various control parameters, making it challenging to maintain quality of service (QoS) and improve resource utilization.
Innovation Solution
A workload manager evaluator is introduced, which employs specific metrics and simulation models to evaluate the performance of workload managers by analyzing control parameter settings, determining optimal values for parameters like minCPU and maxCPU allocation, and assessing resource usage efficiency and QoS, thereby guiding the choice of appropriate parameter values for achieving desirable application QoS while minimizing resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If workload managers use static allocation of resource capacity to consumers, then consumers receive guaranteed resource access, but resource utilization efficiency deteriorates
Solution Approach 1:
The patent implements dynamic allocation of resource capacity by allowing the workload manager to adjust the amount of resource capacity allocated to each consumer based on current workload demands and system conditions, transforming the static allocation model into a dynamic one that adapts to changing conditions while maintaining QoS guarantees
Solution Approach 2:
The patent changes the allocation parameters from fixed static values to dynamic values that can be adjusted based on workload characteristics, resource availability, and performance metrics, enabling the system to optimize both QoS and resource utilization efficiency simultaneously
2Productivity
If workload managers dynamically adjust resource allocation based on workload demand, then resource utilization efficiency improves, but difficulty in evaluating performance and determining optimal control parameters increases
Solution Approach 1:
The patent introduces a feedback mechanism where the workload manager continuously monitors performance metrics and resource utilization, using this feedback information to adjust allocation parameters and control policies, thereby enabling systematic evaluation of performance and determination of optimal parameters through iterative refinement
Solution Approach 2:
The patent introduces an intermediary evaluation layer that mediates between the dynamic allocation process and performance assessment, providing structured metrics and analysis tools that simplify the evaluation of complex dynamic allocation scenarios and help determine optimal control parameters
3Reliability
If consumers are allocated maximum resource capacity, then QoS is satisfied, but resource usage efficiency deteriorates due to over-provisioning
Solution Approach 1:
The patent applies partial action by allocating resource capacity based on actual workload needs rather than providing maximum capacity to all consumers, using demand-based allocation to provide sufficient resources without over-provisioning, thereby improving resource usage efficiency while maintaining QoS satisfaction
Data Source
AI summary
A system comprises a workload manager evaluator operable to receive a representative workload that is representative of competing workloads that share access to at least one shared computing resource. The workload manager evaluator is operable to evaluate performance of a scheduler that schedules access of the competing workloads to the shared computing resource according to defined control parameter values, wherein the workload manager evaluator evaluates performance of the scheduler under the representative workload for a plurality of different values of the control parameters. In certain embodiments, the workload manager evaluator determines an optimal value for the control parameters of the scheduler for scheduling access to the at least one shared computing resource for the representative workload to satisfy defined performance desires of the system.