Endpoint Identifier Segmentation for Cloud-Fog Application Mobility
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network computing architectures face challenges in efficiently determining when to route applications between cloud and fog networks, leading to difficulties in optimizing compute services based on network demand, particularly for real-time processing needs of IoT devices.
Innovation Solution
Implementing a LISP (Locator/Identifier Separator Protocol) architecture that decouples IP addresses from physical locations, allowing endpoint identifiers to remain consistent across different network locations, enabling dynamic routing of packets between cloud and fog networks based on monitored network characteristics and resource availability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If applications are moved between cloud and fog networks based on network demand, then service efficiency and latency are improved, but application reachability and routing complexity worsen
Solution Approach 1:
The patent segments the traditional monolithic IP address into two distinct components: Endpoint Identifier (EID) for application identification and Locator Identifier (RLOC) for physical location. This segmentation allows the EID to remain constant while the RLOC changes based on whether the application runs in cloud or fog network, thereby maintaining reachability while enabling dynamic movement and reducing routing complexity.
Solution Approach 2:
The patent introduces a mapping server as an intermediary component that maintains the mapping between EID and RLOC. This mapping server acts as a mediator that receives packets with EID, looks up the corresponding RLOC, and forwards packets to the correct location, thereby simplifying routing complexity while enabling flexible application movement between cloud and fog networks.
2Adaptability or versatility
If IP addresses are tied to physical locations, then routing is simplified, but application mobility and flexibility between cloud and fog networks worsen
Solution Approach 1:
The patent divides the IP address functionality into two separate identifiers: EID for logical application identification and RLOC for physical network location. This segmentation enables application mobility by allowing the EID to remain constant while the RLOC changes as applications move between cloud and fog networks, thereby improving adaptability while managing addressing complexity through the mapping server.
Solution Approach 2:
The patent implements a dynamic addressing system where the RLOC component can change based on application location (cloud or fog network) while the EID remains static. This dynamic behavior is managed through the mapping server that updates EID-to-RLOC mappings in real-time, enabling flexible application mobility without requiring changes to the application's identity.
3Speed
If applications remain in cloud network, then infrastructure complexity is reduced, but latency and real-time processing capability worsen
Solution Approach 1:
The patent applies local quality by deploying application instances in fog networks that are geographically closer to end-users, thereby reducing latency and improving processing speed for real-time applications. The mapping server directs traffic to the appropriate location (fog or cloud) based on the EID-to-RLOC mapping, enabling fast local processing while maintaining the option for centralized cloud processing when needed.
Solution Approach 2:
The patent implements a dynamic network architecture where applications can be moved between cloud and fog networks based on real-time requirements. The mapping server dynamically updates EID-to-RLOC mappings to reflect current application locations, enabling the system to switch between centralized cloud processing and distributed fog processing, thereby optimizing processing speed while managing infrastructure complexity.
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
Techniques are disclosed for maintaining reachability of an application moving between a central cloud network and a fog network using duplicate endpoint identifiers. Network characteristics of a cloud environment are monitored. The cloud environment includes a central cloud network and a fog network. A server in the central cloud network hosts an application that serves a client device. The application is assigned an endpoint identifier that is mapped to a locator identifier associated with the central cloud network. It is determined that a condition for moving the application to the fog network is satisfied based on the monitored network characteristics. An instance of the application is installed on a server in the fog network. The endpoint identifier is assigned to the application instance and mapped to a locator identifier associated with the fog network.