Content Popularity Ranking in CCN Cache
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current content-centric networks face challenges in efficiently ranking content popularity at high speeds, making it impractical to differentiate between popular and unpopular content objects for caching, especially in high-speed routers that process vast amounts of data.
Innovation Solution
A system that uses self-sampling and exponentially weighted moving averages (EWMA) to determine the popularity of content objects by calculating and updating service rates and system-wide statistics, allowing for the identification of top Nth percentile popular content, and implements a differential object-replacement policy to prioritize popular content in caching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional content caching methods are used in high-speed routers, then all content objects are treated equally, but this makes it impossible to differentiate between popular and unpopular content, leading to inefficient cache utilization
Solution Approach 1:
The patent changes the parameter of content identification from simple presence/absence to a popularity ranking parameter. By calculating service rates and determining popularity levels (popular, neutral, unpopular) based on statistical thresholds, the system transforms equal treatment into differentiated treatment based on measured parameters, thereby improving cache utilization efficiency.
Solution Approach 2:
The system implements self-sampling where content objects themselves provide information about their popularity through service rate measurements. The popularity determination is based on self-generated service rate data compared against system-wide statistics, allowing the cache to automatically differentiate and prioritize content without external intervention, thus improving efficiency while maintaining relatively simple implementation.
2Measurement precision
If service rate calculations are performed for all content objects, then popularity can be determined, but this increases computational complexity and processing time
Solution Approach 1:
The patent applies partial action by focusing computational resources only on determining whether content meets the threshold for popular status rather than precisely ranking all content. The system calculates service rates and compares them against pre-determined statistical thresholds (mean + 2 standard deviations), performing just enough computation to achieve sufficient measurement precision for caching decisions without excessive processing overhead.
Solution Approach 2:
The system performs preliminary calculations of system-wide service rate statistics (mean and standard deviation) before making individual content popularity determinations. By pre-establishing the statistical baseline, the system reduces the computational complexity of individual content assessments to simple threshold comparisons, thereby maintaining measurement precision while minimizing processing time.
3Productivity
If popular content is prioritized in caching, then cache efficiency improves, but this requires complex differential object-replacement policies
Solution Approach 1:
The patent applies local quality by creating different treatment categories (popular, neutral, unpopular) for different content objects based on their service rate characteristics. The differential replacement policy applies specific rules to each category: popular content is protected from eviction, neutral content follows standard LRU, and unpopular content is evicted first. This localized quality differentiation improves cache efficiency while keeping the replacement policy relatively simple through clear categorical rules.
Data Source
AI summary
One embodiment of the present invention provides a system for ranking content popularity in a content-centric network (CCN) content cache. During operation, the system receives an interest in a piece of content stored in the content cache, services the interest by accessing the piece of content, updates a service rate associated with the piece of content, updates system-wide service rate statistics, and determines a popularity level associated with the piece of content based on the updated service rate and the updated system-wide service rate statistics.