WAVE: Popularity-based and collaborative in-network caching for content-oriented networks

Abstract

Abstract-In content-oriented networking, content files are typically cached in network nodes, and hence how to cache content files is crucial for the efficient content delivery and cache storage utilization. In this paper, we propose a content caching scheme, WAVE, in which the number of chunks to be cached is adjusted based on the popularity of the content. In WAVE, an upstream node recommends the number of chunks to be cached at its downstream node, which is exponentially increased as the request count increases. Simulation results reveal that the average hop count of content delivery of WAVE is lower than other schemes, and the inter-ISP traffic volume of WAVE is the second lowest (CDN is the lowest). Also, WAVE achieves higher cache hit ratio and fewer frequent cache replacements than other on-demand caching strategies.

Adaptive Prioritized Probabilistic Caching Algorithm for Content Centric Networks

“…Not only improving the reproduced data quality, but network performance could be enhanced. Because of the uncertainty of content requesting behavior of requesters, we design the new adaptive prioritized probabilistic caching algorithm based on probabilistic caching scheme [5][6][7][8][9][10] and adaptive probabilistic caching scheme [11] which dynamically adapts itself to the network circumstances.…”

“…Several caching schemes were proposed to be used in CCN [3][4][5][6][7][8][9][10][11][12]. Originally, CCN employs universal caching scheme (Always), which allows CS to cache every incoming data packet not existing in the CS.…”

“…The popularity-based caching strategies were presented in [3] and [4], respectively. The probabilistic caching schemes (Prob) were proposed in [5][6][7][8][9][10][11] to randomly cache data packets based on probability. Adaptive probabilistic caching scheme (A-prob) [10] was introduced to adjust caching probability according to recent cache-hit and cache-miss event.…”

“…By nature of data contents mainly multimedia, different portions of data possess different priorities from the compression process. However, the proposed algorithms in [3][4][5][6][7][8][9][10][11] did not consider various priorities and unequal importance of data packets while the prioritized probabilistic caching algorithm [12] cannot maintain the overall network performance yielded by Prob.…”

On the Tradeoff between Performance and User Privacy in Information Centric Networking

“…Another solution for efficient caching is given in [6]. The authors propose WAVE, a collaborative in-network algorithm, where upstream nodes suggest the number of chunks to be stored to their downstream nodes.…”