Edge Video Memory Tiering for Bandwidth-Limited IoT Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Internet-of-things (IoT) applications face bandwidth and backhaul limitations, as well as storage constraints in connected devices and centralized servers, making it challenging to efficiently store and access video data captured by sensors, particularly in systems like intelligent driver monitoring systems and surveillance cameras.
Innovation Solution
Implementing a distributed video storage and search system using edge computing devices with both short-term and long-term memory, where visual data is cached at high resolution for a short period and then compressed or deleted based on priority, allowing for local storage and search of useful visual data, with the option to send relevant data to the cloud for further processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If visual data is stored at high resolution in connected devices, then data quality and accessibility are improved, but storage capacity is quickly exhausted
Solution Approach 1:
The patent segments the storage system into multiple components: connected devices store visual data at high resolution for a limited period, edge computing devices provide intermediate storage capacity, and cloud servers offer long-term archival storage. This segmentation allows each component to operate within its optimal capacity while maintaining high-resolution data accessibility when needed.
Solution Approach 2:
The patent introduces temporal dimension to the storage system by implementing time-based retention policies. High-resolution visual data is maintained in connected devices only for a specific duration (e.g., 24-48 hours), after which it is archived to edge or cloud storage. This temporal segmentation resolves the storage capacity conflict while preserving data quality during the critical retention period.
2Ease of operation
If all visual data is transmitted to centralized servers, then data accessibility is improved, but bandwidth consumption increases significantly
Solution Approach 1:
The patent implements local quality by enabling connected devices to perform preliminary processing and filtering of visual data before transmission. Only relevant or anomalous data is transmitted to edge computing devices or cloud servers, while routine data is processed and stored locally. This selective transmission dramatically reduces bandwidth consumption while maintaining accessibility for important events.
Solution Approach 2:
The patent introduces edge computing devices as intermediaries between connected devices and centralized cloud servers. These edge devices aggregate data from multiple connected devices, perform local analysis, and transmit only essential data to the cloud. This intermediary layer reduces overall bandwidth consumption while improving data accessibility through distributed processing.
3Duration of action of moving object
If connected devices have larger storage capacity, then local data retention is improved, but device complexity and cost increase
Solution Approach 1:
The patent implements preliminary action by pre-establishing data lifecycle management policies that automatically govern when data should be retained locally, archived to edge devices, or transferred to cloud storage. This automated policy-based approach extends local data retention capabilities without requiring complex manual management systems or additional hardware components.
Solution Approach 2:
The patent enables connected devices to self-manage their storage resources by implementing local decision-making algorithms that automatically determine whether to retain, archive, or delete visual data based on predefined criteria such as data age, event importance, and available storage capacity. This self-service approach extends effective data retention without increasing device complexity.
Data Source
AI summary
Systems and methods are provided for distributed video storage and search over edge computing devices having a short-term memory and a long-term memory. The method may comprise caching a first portion of data on a first device. The method may further comprise determining, at a second device, whether the first device has the first portion of data. The determining may be based on whether the first piece of data satisfies a specified criterion. The method may further comprise sending the data, or a portion of the data, and/or a representation of the data from the first device to a third device.