Edge Computing Platform Wireless Mesh Architecture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing edge computing platforms face challenges in bandwidth and latency, particularly in interconnecting edge computing systems, distributing processing power, and managing processing and data storage tasks effectively.
Innovation Solution
A new edge computing platform architecture based on a wireless mesh network with multiple tiers of nodes interconnected via millimeter-wave point-to-point (ptp) or point-to-multipoint (ptmp) links, enabling distributed processing and data storage closer to data generation and consumption sites.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If a centralized, cloud-based computing platform is used for data processing and storage, then data processing capacity is improved, but bandwidth consumption increases and latency increases
Solution Approach 1:
The centralized computing platform is segmented into multiple distributed edge computing systems deployed at different locations (PoP sites, seed sites, anchor sites). Each edge computing system independently processes data locally, eliminating the need to transmit all data to a central cloud platform, thereby reducing bandwidth consumption and latency while maintaining processing capacity.
Solution Approach 2:
The architecture introduces a spatial dimension by distributing computing resources across multiple geographic locations rather than concentrating them in a single central location. This multi-dimensional deployment allows data processing to occur closer to data sources and consumers, reducing transmission distance and latency.
2Power
If a centralized, cloud-based computing platform is used for data processing and storage, then data processing capacity is improved, but bandwidth consumption increases
Solution Approach 1:
The centralized computing platform is segmented into multiple distributed edge computing systems deployed at different locations (PoP sites, seed sites, anchor sites). Each edge computing system independently processes data locally, eliminating the need to transmit all data to a central cloud platform, thereby reducing bandwidth consumption and latency while maintaining processing capacity.
3Loss of time
If existing edge computing platform architecture is used, then some bandwidth and latency improvements are achieved, but processing power distribution remains uniform and inefficient
Solution Approach 1:
The patent implements heterogeneous processing power distribution where different edge computing systems have different levels of processing capability based on their local requirements and resources. PoP sites, seed sites, and anchor sites are configured with appropriate processing power to handle local workloads, eliminating the inefficiency of uniform processing power distribution while reducing complexity through localized decision-making.
Data Source
AI summary
Disclosed herein is an architecture for an edge computing platform based on an underlying wireless mesh network. The architecture includes nodes installed with equipment for operating as part of a wireless mesh network, including (1) a first tier of one or more Point of Presence (PoP) node, (2) a second tier of one or more seed nodes that are each directly connected to at least one PoP node via a PoP-to-seed wireless link, and (3) a third tier of one or more anchor nodes that are each connected to at least one seed node either (i) directly via a seed-to-anchor wireless link or (ii) indirectly via one or more intermediate anchor nodes, one or more anchor-to-anchor wireless links, and one seed-to-anchor wireless link, where at least one node in each of these tiers is further installed with equipment for operating as part of an edge computing platform.