Flexible Instance Scheduling Under QoS and Utilization Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The low utilization of servers in cloud data centers, typically ranging from 10% to 20%, leads to high equipment and operational costs, increased power consumption, and environmental impact, necessitating improved resource management.
Innovation Solution
A cloud data center with a resource management system that monitors and adjusts the size of flexible instances based on QoS guarantees, resource utilization thresholds, and specifications to optimize resource allocation and utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the size of flexible instances is adjusted based on resource utilization to improve server utilization rate, then resource utilization improves, but service stability may deteriorate
Solution Approach 1:
The patent implements dynamic adjustment of flexible instance sizes based on real-time resource utilization monitoring. The resource management system continuously monitors CPU utilization, memory utilization, and I/O utilization metrics, and adjusts instance sizes dynamically within specified ranges to optimize server utilization while maintaining service stability through controlled adaptation rather than static allocation
Solution Approach 2:
The system changes physical parameters of flexible instances (CPU cores, memory size, storage capacity) based on monitored resource utilization thresholds. When utilization exceeds upper thresholds, instance sizes are reduced; when below lower thresholds, sizes are increased, thereby adjusting system parameters to resolve the contradiction between high utilization and service stability
2Productivity
If the size of flexible instances is reduced to improve resource utilization, then resource utilization improves, but QoS guarantee may deteriorate
Solution Approach 1:
The resource management system implements feedback control by continuously monitoring resource utilization metrics and adjusting flexible instance sizes accordingly. The system uses upper and lower threshold feedback mechanisms: when utilization exceeds the upper threshold, instances are shrunk; when below the lower threshold, instances are expanded. This closed-loop feedback ensures QoS guarantees are maintained while optimizing resource utilization
Solution Approach 2:
The system performs preliminary actions by pre-configuring upper and lower utilization thresholds and size adjustment ranges before resource contention occurs. The resource management system is pre-programmed with QoS policies and adjustment rules that are activated automatically when threshold conditions are met, preventing QoS degradation before it occurs rather than reacting after degradation has happened
Data Source
Figure 1
Figure 2
Figure 3
AI summary
This application provides a cloud data center, including a resource management system and a computing resource pool. The resource management system monitors a running parameter of a flexible instance running in the computing resource pool, and indicates, according to an adjustment requirement, the computing resource pool to adjust a size of the flexible instance, where the adjustment requirement includes a QoS guarantee requirement and the running parameter of the flexible instance. The computing resource pool runs the flexible instance, collects the running parameter of the flexible instance, and adjusts, based on the indication, the size of the flexible instance, where a process of adjusting meets the QoS guarantee requirement. The cloud data center provides running of a flexible instance, improves a resource utilization of the cloud data center, reduces energy consumption, and generates environmental protection benefits.