Virtual Instance Resource Pooling Over High-Speed Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The performance of virtual instances in serverless computing is not fully utilized in separate data centers, leading to high operating costs.
Innovation Solution
A cloud computing technology-based virtual instance configuration method that dynamically configures and adjusts computing, memory, storage, and network resources using a high-speed network-connected infrastructure comprising computing, memory, and network pools, allowing for resource-intensive cloud-native resource pools with low-cost and high-performance virtual instances.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If virtual instance operating mode is directly applied to separate data center, then implementation simplicity is improved, but resource utilization performance deteriorates
Solution Approach 1:
The patent implements dynamic resource allocation by introducing a resource scheduling module that continuously monitors virtual instance performance and adjusts resource allocation in real-time. The system dynamically selects from multiple resource allocation strategies (static allocation, dynamic allocation, on-demand allocation) based on current workload conditions, transforming the rigid direct-application approach into a flexible adaptive system that optimizes resource utilization while maintaining implementation feasibility.
Solution Approach 2:
The system changes key operational parameters by implementing multiple resource allocation strategies with different parameter configurations. The scheduling module adjusts allocation parameters (CPU quotas, memory limits, storage I/O rates) based on virtual instance performance metrics and workload characteristics, enabling the system to adapt parameters dynamically rather than using fixed direct-application settings.
2Reliability
If high-performance operating resources are fixedly configured, then performance guarantee is improved, but operating cost deteriorates
Solution Approach 1:
The patent applies partial action by allocating resources based on actual needs rather than providing full high-performance resources continuously. The resource scheduling module implements quota-based allocation where virtual instances receive sufficient resources to meet performance requirements without excessive allocation, reducing operating costs while maintaining necessary performance guarantees through dynamic adjustment.
Solution Approach 2:
The system implements feedback mechanisms by monitoring virtual instance performance metrics and using this information to adjust resource allocation. The scheduling module receives performance feedback from virtual instances and modifies resource allocation strategies accordingly, ensuring performance guarantees are maintained while avoiding unnecessary resource consumption and associated costs.
3Adaptability or versatility
If resources are allocated from separate pools via high-speed network, then resource flexibility is improved, but system complexity deteriorates
Solution Approach 1:
The patent introduces a resource scheduling module as an intermediary between the separate resource pools and virtual instances. This intermediary manages the complexity of allocating resources from computing pools, storage pools, and network pools by implementing standardized allocation protocols and abstraction layers, thereby maintaining resource flexibility while controlling system complexity through centralized coordination.
Solution Approach 2:
The resource scheduling module serves multiple functions simultaneously: it allocates resources from different pools, monitors performance metrics, adjusts allocation strategies, and manages multiple virtual instances. This multi-functional intermediary handles the complexity of flexible resource allocation from separate pools through a single unified system that manages diverse resource types and allocation scenarios.
Data Source
Figure 1
Figure 2
Figure 3~5
AI summary
Embodiments of this application disclose a cloud computing technology-based virtual instance configuration method and a cloud management platform, and relate to the field of computer technologies. The method includes: obtaining a virtual instance creation request input by a tenant; and creating a virtual instance in a first server of a computing pool based on the virtual instance creation request. The virtual instance creation request carries specification configuration information of the virtual instance, and the specification configuration information indicates a computing resource, a memory resource, a storage resource, and a network resource that are required by the virtual instance. The computing resource of the virtual instance is provided by the first server, the memory resource of the virtual instance is provided by a second server of a memory pool for the virtual instance in the first server via a high-speed network, the storage resource of the virtual instance is provided by a third server of a storage pool for the virtual instance in the first server via the high-speed network, and the network resource of the virtual instance is provided by a fourth server of a network pool for the virtual instance in the first server via the high-speed network.