Cloud Resource Request Routing Across Adjacent Data Centers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Centralized management of resource allocation requests in large-scale cloud computing systems is complex and delays resource allocation, particularly when numerous data centers are involved, leading to suboptimal service quality due to high latency and network congestion.
Innovation Solution
A method and device for managing computing resource allocation in a cloud system by sequentially routing allocation requests to adjacent data centers along a predefined path, starting from the closest data center to the user, until a resource is available or all centers are contacted, optimizing the allocation process and minimizing network congestion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If centralized management of resource allocation requests is implemented in large-scale cloud computing systems, then resource allocation can be coordinated across multiple data centers, but the complexity of management increases and resource allocation is delayed
Solution Approach 1:
The patent segments the centralized management function into distributed components at each data center. Each data center independently manages its own resource allocation requests using local policies, rather than funneling all requests through a central management system. This segmentation reduces management complexity while maintaining coordination across the cloud computing system through standardized request handling procedures.
2Loss of time
If the number of data centers is increased to reduce latency for users, then service quality improves, but the complexity of managing allocation requests increases and network congestion worsens
Solution Approach 1:
The patent applies local quality by enabling each data center to independently evaluate and process allocation requests according to its local resource availability and policies. This local decision-making capability allows data centers to respond quickly to user requests without centralized coordination overhead, reducing latency while managing complexity at the local level rather than system-wide.
3Adaptability or versatility
If allocation requests are sequentially routed to multiple data centers to find available resources, then resource availability increases, but network congestion increases and allocation time increases
Solution Approach 1:
The patent implements preliminary action by having data centers pre-evaluate their resource availability and publish this information in advance. When allocation requests are received, data centers can quickly determine if they can fulfill the request based on pre-assessed availability, rather than sequentially querying each data center. This preliminary preparation reduces both allocation time and network congestion while maintaining high resource availability.
Data Source
Figure 1~2
Figure 3~4
AI summary
A method for managing a request to allocate a computing resource in a cloud computing system (1) comprising at least two data centers (DC1, DC2, DC3) connected by a communication network (200) is implemented by an access device (20) allowing a terminal (10) to access the communication network and determining a routing path (300) to a service address. The method includes: - the transmission (E10) of the request (Req) to a first data center (DC1); and - if the computing resource is not available, the retransmission (E11) of the request (Req) to an adjacent data center that is next in the routing path (300), the retransmission (E12) of the request (Req) being withdrawn until either a data center responds that the resource is available or the request (Req) has been retransmitted to all data centers.