Edge-Based Resource Spin-Up for Cloud Latency Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud computing systems face performance issues due to centralized resource provisioning, leading to delayed responses and poor user experiences, especially for users geographically distant from data centers, and lack flexibility in rerouting requests during cloud provider difficulties.
Innovation Solution
Implementing edge-based resource spin-up within cloud computing environments, where virtualized instances are created near user request sources, and dynamically routing requests to multiple cloud environments based on specified criteria to ensure quality of service and efficient performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If resources are spun up in a centralized cluster, then resource management is simplified, but user response time deteriorates for users located far from the data center
Solution Approach 1:
The patent segments the centralized cloud resources into multiple edge locations distributed geographically closer to users. Each edge location can independently spin up resources locally, eliminating the single centralized cluster while maintaining manageable resource allocation through distributed control planes.
Solution Approach 2:
The patent introduces a spatial dimension to resource deployment by distributing compute resources across multiple geographic locations (edge data centers) rather than concentrating them in a single data center. This dimensional change enables users to access resources from nearby edge locations, significantly reducing latency while maintaining simplified management through orchestration layers.
2Device complexity
If applications are deployed to a single cloud, then deployment complexity is reduced, but service reliability deteriorates when the cloud provider experiences difficulties
Solution Approach 1:
The patent segments the monolithic cloud deployment into multiple independent cloud environments (edge clouds) distributed across different geographic locations and potentially different providers. Each edge cloud can operate independently, allowing requests to be routed to functional clouds even when others experience difficulties, thereby improving reliability without significantly increasing deployment complexity through standardized edge cloud templates.
Solution Approach 2:
The patent changes the parameter of cloud deployment from single-location to multi-location distribution. By deploying applications across multiple edge clouds with different geographic and provider parameters, the system achieves improved reliability and fault tolerance while maintaining manageable deployment complexity through parameterized deployment configurations.
3Ease of manufacture
If compute resources are located far from users, then infrastructure consolidation is improved, but computational performance deteriorates due to routing delays
Solution Approach 1:
The patent resolves this contradiction by adding a spatial distribution dimension to the consolidated infrastructure. Instead of consolidating all resources in one distant data center, the infrastructure is consolidated across multiple edge locations geographically distributed near users. This maintains the consolidation benefits of standardized management while eliminating the performance penalty of distant resource location through proximal edge deployment.
Solution Approach 2:
The patent segments the consolidated infrastructure into distributed edge compute resources located near users. Each edge location maintains consolidated resource management locally while the overall system achieves infrastructure consolidation through standardized edge cloud deployments. This segmentation enables computational performance improvement through local resource access while preserving infrastructure consolidation benefits.
Data Source
AI summary
Aspects of the present invention include distributing new resources closer to end-users which are making increased demands by spinning-up additional virtualized instances (as part of a cloud provisioning) within servers that are physically near to the network equipment (i.e., web servers, switches, routers, load balancers) that are receiving the requests.