Cloud DVR Pre-fetching Content to Reduce DSN Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed storage networks (DSNs) used in cloud-based Digital Video Recorders (DVRs) face inefficiencies due to peak and valley periods in content demand, leading to inefficient resource allocation and potential delays in servicing user requests.
Innovation Solution
Implementing a cloud-based DVR that pre-fetches content from DSN memory during low utilization periods if the content has a high estimated probability of being requested, based on historical access patterns, popularity, and future recording or viewing markers, and caches it locally to minimize peak load impacts on the DSN.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If content is pre-fetched from DSN memory during low utilization periods, then the load on DSN during peak periods is reduced, but additional processing and storage resources are consumed at cloud-based DVRs
Solution Approach 1:
The system performs preliminary actions by pre-fetching content from DSN memory to local cache during low-utilization periods. The cloud-based DVR monitors DSN utilization and proactively retrieves likely-to-be-requested content before peak demand occurs, reducing peak load on the distributed storage network.
Solution Approach 2:
A local cache acts as an intermediary between the cloud-based DVR and the DSN memory. This intermediate storage layer buffers content requests, allowing the system to serve frequently accessed content locally during peak periods without directly burdening the DSN, thus reducing overall system complexity while maintaining productivity.
2Speed
If content is cached locally at cloud-based DVR, then access speed during peak periods is improved, but memory resources at DVR are consumed
Solution Approach 1:
The system applies local quality by caching content selectively at individual cloud-based DVRs based on their specific access patterns and local demand characteristics. Each DVR maintains a customized cache tailored to its user base, optimizing local access speed while consuming only the necessary memory resources for its specific workload.
Solution Approach 2:
The system dynamically adjusts caching parameters such as cache size, pre-fetch thresholds, and retention policies based on monitored conditions including DSN utilization levels, local access patterns, and available memory resources. This allows optimization of access speed while adapting memory consumption to actual system state.
3Measurement precision
If pre-fetching is performed based on historical access patterns, then accuracy of predicting user demand is improved, but data processing overhead increases
Solution Approach 1:
The system performs partial analysis by focusing on the most significant access patterns and content types when predicting demand. Rather than analyzing all possible data points, it concentrates processing on key indicators such as historically frequent access patterns and high-demand content categories, achieving sufficient prediction accuracy with reduced processing overhead.
Data Source
AI summary
A processing device implementing a cloud-based digital video recorder (DVR) or providing cloud-based streaming services includes a local memory that can be used for storing content prior to providing the content to requestors. The cloud-based DVR determines an estimated probability that delivery of a particular content item will be requested within a particular period of time, and a difference between current and average loads of a second memory included in another DVR. The cloud-based DVR pre-fetches part or all of the particular content item from other DVR's memory into the local memory based on the estimated probability that delivery of the particular content item will be requested within the particular period of time, and the difference between the current and average loads on the other DVR's memory.