Multi-cluster Resource Management via Virtual Access Point Network Map
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Managing resources in a multi-cluster computing environment from a centralized access point is challenging due to the need for replicating large amounts of entity data, which consumes significant storage and network resources, especially in highly dynamic environments.
Innovation Solution
Implementing a communication and processing regime that uses a virtual access point network map to route resource management requests between computing clusters, reducing the demand for computer memory, processing power, and network bandwidth by identifying and routing requests through a network of access points determined by a network map of inter-cluster connections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If entity data is replicated at the centralized access point, then resource management functionality is provided, but storage resources and network I/O resources are consumed significantly
Solution Approach 1:
The patent extracts the entity data from the centralized access point and keeps it only at the edge access points where the resources actually reside. The centralized access point maintains only routing information and access point network map, while edge access points store the actual entity data locally, eliminating the need for large-scale data replication.
Solution Approach 2:
The patent introduces edge access points as intermediaries between the centralized access point and the resources. These edge access points act as local proxies that store entity data and handle resource management operations locally, reducing the need for data replication to the centralized access point while still providing centralized coordination through the access point network map.
2Ease of operation
If entity data is replicated at the centralized access point, then resource management functionality is provided, but network I/O resources are consumed significantly
Solution Approach 1:
The patent extracts entity data from the centralized access point storage requirements and relocates it to edge access points. This eliminates the need for continuous network I/O for data replication to the centralized access point, as data remains locally stored at edge access points closer to the resources.
Solution Approach 2:
The patent changes the architectural dimension from a single centralized access point to a hierarchical multi-dimensional structure with centralized and edge access points. This dimensional change allows data to be stored at the edge dimension closest to resources, eliminating the need for network I/O to replicate data to the centralized dimension.
3Ease of operation
If large amounts of entity data are replicated, then resource management functionality is provided, but resource consumption increases over time
Solution Approach 1:
The patent extracts entity data from the centralized access point and stores it locally at edge access points. This eliminates the growing resource consumption problem because data is not replicated to the centralized access point, so resource usage remains constant over time rather than increasing with the number of managed resources.
Solution Approach 2:
Edge access points serve themselves by storing entity data locally and handling resource management operations independently. This self-service capability eliminates the need for continuous data replication to the centralized access point, maintaining constant resource consumption regardless of the number of resources managed.
Data Source
AI summary
Entries in individual portions of a network map serve to establish inter-cluster routes that are formed from routing data structures at individual clusters, which routing data structures store connection parameters pertaining to access point nodes in a multi-cluster computing environment. Upon receiving a resource management request at one of the access points, the request parameters are used in combination with the network map to identify at least one target access point having a virtual machine for responding to the resource management request. The resource management request is routed to a next hop access point node on route(s) to target access points. If the next hop access point node is not the node to respond to the resource management request, then processing at the next hop node serves to route the resource management request to a further next hop access point node on a route to the target access point.