# PopNetCod: A Popularity-based Caching Policy for Network Coding enabled   Named Data Networking

**Authors:** Jonnahtan Saltarin, Torsten Braun, Eirina Bourtsoulatze, and Nikolaos, Thomos

arXiv: 1901.01187 · 2019-01-07

## TL;DR

PopNetCod is a distributed, popularity-based caching policy for network coding enabled Named Data Networking that improves cache-hit rates and network performance by caching the most popular data closer to users.

## Contribution

It introduces a novel distributed caching policy leveraging content popularity and network coding, enhancing cache efficiency in Named Data Networking.

## Key findings

- Higher cache-hit rate compared to Leave Copy Everywhere and LRU policies
- Increased goodput for clients due to better caching
- Reduced load on source servers

## Abstract

In this paper, we propose PopNetCod, a popularity-based caching policy for network coding enabled Named Data Networking. PopNetCod is a distributed caching policy, in which each router measures the local popularity of the content objects by analyzing the requests that it receives. It then uses this information to decide which Data packets to cache or evict from its content store. Since network coding is used, partial caching of content objects is supported, which facilitates the management of the content store. The routers decide the Data packets that they cache or evict in an online manner when they receive requests for Data packets. This allows the most popular Data packets to be cached closer to the network edges. The evaluation of PopNetCod shows an improved cache-hit rate compared to the widely used Leave Copy Everywhere placement policy and the Least Recently Used eviction policy. The improved cache-hit rate helps the clients to achieve higher goodput, while it also reduces the load on the source servers.

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/1901.01187/full.md

## Figures

11 figures with captions in the complete paper: https://tomesphere.com/paper/1901.01187/full.md

## References

24 references — full list in the complete paper: https://tomesphere.com/paper/1901.01187/full.md

---
Source: https://tomesphere.com/paper/1901.01187