Dynamic Application Server Clustering via QoS History
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing application server clusters face inefficiencies during heavy workloads, as the number of servers remains constant, and the replacement server may not be optimally suited to handle the workload, leading to potential performance issues and resource misallocation.
Innovation Solution
A system that dynamically selects the best-suited application server by maintaining a history of quality of service (QoS) and adding it to the cluster when workload exceeds a threshold, releasing it when the workload decreases, ensuring optimal resource allocation and performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the number of servers in the cluster is increased to handle heavy workload, then the handling capacity is improved, but the resource allocation efficiency deteriorates due to idle servers during light workload periods
Solution Approach 1:
The cluster size is made dynamic rather than fixed. The system automatically adjusts the number of servers in the cluster based on real-time workload conditions, expanding during heavy workload and contracting during light workload periods. This resolves the contradiction by allowing the system to have high handling capacity when needed while maintaining resource allocation efficiency when workload is low.
Solution Approach 2:
The system changes the parameter of cluster size based on workload conditions. By monitoring workload metrics and adjusting the number of active servers accordingly, the system optimizes both handling capacity and resource allocation efficiency at different operational states.
2Productivity
If a replacement server is added to the cluster during heavy workload, then the handling capacity is improved, but the performance reliability deteriorates because the replacement server may not be optimally suited for the workload
Solution Approach 1:
The system performs preliminary evaluation and selection of replacement servers before they are added to the cluster. By assessing server suitability in advance based on workload requirements and server capabilities, the system ensures that only appropriately matched servers are incorporated, maintaining performance reliability while expanding handling capacity.
Solution Approach 2:
The system implements feedback mechanisms to monitor the performance and suitability of added servers. This continuous feedback allows the system to verify that replacement servers are performing as expected and to make adjustments if performance requirements are not met, thereby maintaining reliability.
3Loss of energy
If policy-based dynamic allocation of resources is used to assign fewer resources to idle servers, then the resource allocation efficiency is improved, but the handling capacity remains limited due to constant cluster size
Solution Approach 1:
The system transitions from static cluster sizing to dynamic cluster sizing. While policy-based dynamic allocation optimizes resource distribution within a fixed cluster, this invention additionally dynamically adjusts the cluster size itself, allowing the system to expand handling capacity when needed while maintaining resource efficiency through intelligent server addition and removal.
Data Source
AI summary
A system, computer program product and computer-implemented method of dynamically adding a best suited application server to a cluster of application servers are provided. To do so, a history of quality of service (QoS) of a plurality of application servers is maintained. Using the history of QoS, an application server is dynamically selected and added to the cluster when the cluster is under a heavy workload. The selected application server is one that has the most historically favorable QoS and a heavy workload is one that exceeds a user-configurable threshold.