Employing Working Set Model for the Content Store in Content-Centric Network
Seung‐Seok Kang, Longzhe Han, Matt W. Mutka · Advanced science and technology letters · 2014
Content-Centric Network (CCN) is one of the future Internet architecture models in which data can be retrieved using its name rather than a server address. In addition, CCN routing nodes perform content caching and multicast transmission. The node contains pending interest table (PIT), content store (CS), and forwarding information base (FIB) to efficiently deliver data packets. Clients may send an interest packet in order to request a corresponding data packet. The content server or any intermediate node that caches the data responds by transmitting the data. When several clients request the same content, the caching in CS helps improving the transmission performance of the content flow. If an intermediate node allocates a proper memory size in CS for the content flow, it increases the transmission performance of the content flow. When the memory size for the flow is less than the appropriate size, it suffers from degrading the performance. This paper proposes the necessity of maintaining the suitable size in CS, called a working set size, for a content flow, as well as deciding the proper replacement algorithm.