Fine-grained cache deployment scheme for arbitrary topology in ICN

Duan, Jie; Xing, Yuan; Zeng, Shuai; Liu, Yuanni

doi:10.1007/s00521-017-3288-x

Cited by 2 publications

(5 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…So the overall complexity of our allocation method is max(O |M| 2 , O(T budget log(T budget ). The computational complexity of optimal allocation method in Reference [20] is max(O |M| 3 , O(T budget log(|C| ). Compared with the optimal allocation methods, the computational complexity of our method is relatively low.…”

Section: Methods Descriptionmentioning

confidence: 99%

“…Heterogeneous cache allocation schemes have been the main part of some studies such as [7,[18][19][20]. The authors of Reference [7] studied graph-related centralities metrics and they noticed that the gain brought by the heterogeneous cache is very limited.…”

Section: Cache Resource Allocationmentioning

confidence: 99%

“…Apart from considering topology only, References [19,20] also took network traffic into account. In Reference [19] cache was allocated according to short-term parameters such as user behaviors and outgoing traffic as well as network topology.…”

Section: Cache Resource Allocationmentioning

confidence: 99%

“…This method is suitable for networks where content changes dynamically but cannot allocate cache properly before gathering enough traffic information. Duan et al proposed a fine-grained cache deployment scheme to allocate cache [20], while fine grain size should be chosen properly. Besides, References [8,9] took content popularity into consideration.…”

Section: Cache Resource Allocationmentioning

confidence: 99%

“…If a request arrives at a node that contains the requested content, the requested data will be sent from the hit node rather than from the origin server. This three-layer network can be easily extended to a larger multilayer tree topology, such as doing k-core decomposition for networks in Reference [20]. We assume that all users are connected to the leaf nodes and their requests for content will arrive at leaf nodes first.…”

Section: Model Descriptionmentioning

confidence: 99%

See 4 more Smart Citations

An On-Path Caching Scheme Based on the Expected Number of Copies in Information-Centric Networks

Wang

Han

2020

Electronics

View full text Add to dashboard Cite

The Information-Centric Network (ICN) is one of the most influential future network architectures and in-network caching in ICN brings some helpful features, such as low latency and mobility support. How to allocate cache capacity and place content properly will greatly influence the performance of ICN. This paper focuses on the cache allocation problem and content placement problem under the given cache space budget. Firstly, a lightweight allocation method utilizing information of both topology and content popularity is proposed, to allocate cache space and get the expected number of copies of popular content. The expected number of copies represents the number of content copies placed in the topology. Then, an on-path caching scheme based on the expected number of copies is proposed to handle the content placement problem. In the cache allocation scenario, the lightweight allocation method performs better than other baseline methods. In the content placement scenario, Leave Copy Down (LCD) based on the expected number of copies performs the second-best and is very close to Optimal Content Placement (OCP).

show abstract

Section: Methods Descriptionmentioning

confidence: 99%