Combined Network and Computation Slicing for Edge Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current edge computing solutions, such as Mobile Cloud Computing, introduce significant communication delays due to the remote location of cloud servers, making them unsuitable for real-time applications, and existing network slicing technologies do not effectively account for computational resources in ensuring quality of service for latency-critical applications.
Innovation Solution
The implementation of combined network and computation slicing (NCS) mechanisms, which involve a controller that reserves both network and computational resources, including Multi-access Edge Computing (MEC) resources, to create a network and computational slicing instance (NCSI) that dynamically monitors and adjusts operational parameters to ensure quality of service (QoS) across both communication and computational layers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If computational offloading is performed to centralized cloud servers, then computational capacity is improved, but communication latency increases significantly
Solution Approach 1:
The patent segments the centralized cloud computing system into distributed edge computing nodes located at network edges closer to users. This segmentation allows computational tasks to be executed locally at edge nodes rather than being centralized, thereby reducing communication latency while maintaining computational capacity.
Solution Approach 2:
The patent introduces a spatial dimension change by moving computational resources from centralized data centers to distributed edge locations throughout the network. This dimensional shift from centralization to distribution enables simultaneous improvement of both computational capacity and reduction of communication latency by placing compute resources physically closer to data sources and consumers.
2Ease of operation
If network slicing is implemented separately from computation slicing, then network resource management is improved, but quality of service for latency-critical applications deteriorates
Solution Approach 1:
The patent merges network slicing and computation slicing into a unified framework where network resources and computational resources are jointly allocated and managed. This integration ensures that network slices are directly coupled with appropriate edge computing resources, enabling end-to-end QoS guarantees for latency-critical applications by coordinating both network and compute resource allocation.
Solution Approach 2:
The patent creates a universal resource management framework that simultaneously handles both network resources and computational resources. This multi-functional system enables a single slicing mechanism to manage diverse resource types (network bandwidth, compute power, storage) in a coordinated manner, improving overall QoS for latency-sensitive applications.
Data Source
AI summary
Methods and devices for creating and operating a combined network and computational slice instance (NCSI) in a Multi-access Edge Computing (MEC) scenario. Communication and computational resources may be reserved by a NCSI controller for the NCSI. The communication resources may include network slices and the computational resources may include MEC computational resources of one or more MEC servers. The reserved resources may be selected based on quality of service (QoS) requirements of UEs that will utilize the NCSI. During operation, reserved resources for the NCSI may be dynamically renegotiated based on an aggregate load of the NCSI, the QoS of data traffic, and/or updated QoS requirements of the UEs.