Cloud Orchestrator Hierarchical Resource Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud network technologies lack a unified, rule-based algorithm for orchestrating and relocating computing resources across multiple datacenters to ensure resilience against failures, particularly in cases of multiple failures in local and geographically distributed datacenters.
Innovation Solution
A cloud orchestrator system that computes a resource map with global and regional tiers, compares resource needs with the map to determine allocation solutions, and allocates resources across datacenters based on defined rules for resiliency requirements, enabling dynamic relocation and rebalancing of network functions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If computing resources are allocated across multiple geographically distributed datacenters to ensure resilience, then reliability is improved, but device complexity increases due to the need for hierarchical resource mapping and rule-based orchestration
Solution Approach 1:
The patent segments the cloud network resources into hierarchical tiers (global tier and regional tier) to manage complexity. The cloud orchestrator computes a resource map that organizes computing resources across multiple datacenters into structured levels, allowing resilient allocation without overwhelming system complexity. This segmentation enables the system to handle geographically distributed resources systematically.
Solution Approach 2:
The cloud orchestrator performs preliminary actions by pre-computing the resource map and establishing rule-based algorithms before failures occur. The system proactively determines allocation solutions based on resiliency requirements and operator-defined rules, so that when failures happen, resources can be rapidly reallocated without complex real-time decision-making under pressure.
2Productivity
If a rule-based algorithm is implemented to prioritize relocation of reserved computing resources, then productivity is improved through efficient resource allocation, but device complexity increases due to the orchestration requirements
Solution Approach 1:
The patent applies parameter changes by using resiliency scores as a key parameter to prioritize resource relocation. The cloud orchestrator evaluates computing resources based on resiliency requirements and operator-defined rules, transforming the allocation decision into a parameter-driven process. This allows efficient prioritization of which resources to relocate first without requiring complex manual orchestration.
Solution Approach 2:
The system creates a virtual copy of the resource allocation state through the computed resource map. This abstract representation allows the orchestrator to simulate and evaluate different allocation scenarios before implementing changes, improving productivity by avoiding trial-and-error reallocations while managing complexity through simulation rather than direct manipulation.
3Reliability
If resources are allocated based on resiliency requirements and operator rules, then reliability is improved, but loss of time increases due to the computation and comparison processes
Solution Approach 1:
The cloud orchestrator performs preliminary computation of the resource map and pre-establishes rule-based algorithms for resource allocation. By having the resource hierarchy and allocation rules predetermined, the system minimizes decision time when failures occur, as the orchestrator can quickly match failed resources against the pre-computed map and rules rather than creating allocation strategies from scratch under pressure.
Solution Approach 2:
The system enables self-service through automated rule-based allocation that operates without manual intervention. The cloud orchestrator autonomously compares resource needs with the resource map, determines allocation solutions, and relocates resources based on resiliency requirements and operator-defined rules, reducing the time loss associated with manual orchestration while maintaining high reliability.
Data Source
AI summary
A method includes receiving a request to allocate an instantiation of a network function and information indicative of resource needs of the instantiation. The resource needs include at least one resiliency requirement. The method includes computing a resource map comprising a global tier and a regional tier and comparing the resource needs with the resource map to determine an allocation solution. The method also includes, based on the allocation solution, allocating resources to the instantiation. The resources include a first resource of the global tier and a second resource of the regional tier.