Intelligent Web proxy caching approaches based on machine learning techniques

Ali, Waleed; Shamsuddin, Siti Mariyam; Ismail, Abdul Samad

doi:10.1016/j.dss.2012.04.011

Cited by 53 publications

(60 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When the objects are moved from short term cache to long term cache SVM classifier is used to classify the Web objects as class 0 or class 1 [2]. When a request is made for a Web object, Improved Performance by Combining Web Pre-Fetching Using Clustering with Web Caching Based on SVM Learning Method 169…”

Section: The Proposed Methods Of Combined Clustering Based Pre-fetchinmentioning

confidence: 99%

“…Baskaran, C. Kalaiarasan <a1, a2, a3, a4, b> where a1 represents the recency of the Web object, a2 represents frequency of the Web object, a3 represents the size of the Web object, a4 represents the retrieval time (access latency) of the Web object and b represents the class of the Web object. The data type of a1, a2, a3, a4 is numeric and the data type of b is nominal [2].…”

Section: Preprocessing Step and Creation Of Training Datasetmentioning

confidence: 99%

“…The traditional cache replacement techniques used does not increase the cache hit ratio to a great extent. Machine learning techniques are deployed to improve the performance of Web proxy caching methods [2]. Compared to traditional caching approaches, intelligent Web caching methods are more efficient.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Improved Performance by Combining Web Pre-Fetching Using Clustering with Web Caching Based on SVM Learning Method

Baskaran¹,

Chellan²

2016

INT J COMPUT COMMUN

View full text Add to dashboard Cite

Combining Web caching and Web pre-fetching results in improving the bandwidth utilization, reducing the load on the origin server and reducing the delay incurred in accessing information. Web pre-fetching is the process of fetching the Web objects from the origin server which has more likelihood of being used in future. The fetched contents are stored in the cache. Web caching is the process of storing the popular objects "closer" to the user so that they can be retrieved faster. In the literature many interesting works have been carried out separately for Web caching and Web pre-fetching. In this work, clustering technique is used for pre-fetching and SVM-LRU technique for Web caching and the performance is measured in terms of Hit Ratio (HR) and Byte Hit Ratio (BHR). With the help of real data, it is demonstrated that the above approach is superior to the method of combining clustering based prefetching technique with traditional LRU page replacement method for Web caching.

show abstract

Section: The Proposed Methods Of Combined Clustering Based Pre-fetchinmentioning

confidence: 99%

Section: Preprocessing Step and Creation Of Training Datasetmentioning

confidence: 99%

See 1 more Smart Citation

Improved Performance by Combining Web Pre-Fetching Using Clustering with Web Caching Based on SVM Learning Method

Baskaran¹,

Chellan²

2016

INT J COMPUT COMMUN

View full text Add to dashboard Cite

show abstract

“…Byte hit rate measures the amount of data (in bytes) served from the cache as a percentage of the total amount of bytes requested. These metrics are among the most commonly used to evaluate caching systems [9,[11][12][13]27], and allow us to analyze the ability of our caching system in caching the pages that are most likely to be requested in the near future.…”

Section: Simulation Environmentmentioning

confidence: 99%

“…At the other end of the spectrum, more complex solutions based on machine learning try to predict which pages are going to be more frequently accessed by users based on past behavior [11][12][13]. In between, several other cache replacement algorithms have been proposed [14][15][16][17][18][19][20][21].…”

Section: Introductionmentioning

confidence: 99%

An adaptive semantics-aware replacement algorithm for web caching

Negrão

Roque

Ferreira

et al. 2015

J Internet Serv Appl

View full text Add to dashboard Cite

As the web expands its overwhelming presence in our daily lives, the pressure to improve the performance of web servers increases. An essential optimization technique that enables Internet-scale web servers to service clients more efficiently and with lower resource demands consists in caching requested web objects on intermediate cache servers. At the core of the cache server operation is the replacement algorithm, which is in charge of selecting, according to a cache replacement policy, the cached pages that should be removed in order to make space for new pages. Traditional replacement policies used in practice take advantage of temporal reference locality by removing the least recently/frequently requested pages from the cache. In this paper we propose a new solution that adds a spatial dimension to the cache replacement process. Our solution is motivated by the observation that users typically browse the Web by successively following the links on the web pages they visit. Our system, called SACS, measures the distance between objects in terms of the number of links necessary to navigate from one object to another. Then, when replacement takes place, objects that are distant from the most recently accessed pages are candidates for removal; the closest an object is to a recently accessed page, the less likely it is to be evicted. We have implemented a cache server using SACS and evaluated our solution against other cache replacement strategies. In this paper we present the details of the design and implementation of SACS and discuss the evaluation results obtained.

show abstract