CLUSTERING BASED WEB PREFETCHING IN HIGH TRAFFIC ENVIRONMENT
Need help with a related project topic or New topic? Send Us Your Topic
DOWNLOAD THE COMPLETE PROJECT MATERIAL
CLUSTERING BASED WEB PREFETCHING IN HIGH TRAFFIC ENVIRONMENT
Chapter One: Introduction
1.1 Background of the Study
The web is a collection of text documents and other resources linked by hyperlinks and Uniform Resource Locators (URLs), which are often accessible by web browsers via web servers. The web began as a simple information sharing medium and has since expanded into a diverse collection of dynamic and interactive services.
The huge growth of the web has resulted in a significant demand for bandwidth and a delay in retrieving user requests (Neha, 2013). Users may suffer unexplained delays when getting web pages from the server. Increased bandwidth is one possible solution to the problem, but it has a large economic cost.
Web caching minimises the user’s perceived latency, bandwidth utilisation, and burden on the origin servers (Pallis 2007). Latency is the time elapsed between the moment a request is issued and the time the sender receives the required information.
Many latency-tolerant strategies have been developed throughout the years to address this issue without requiring additional bandwidth. Most notably, caching and prefetching.
Web prefetching helps to fetch and cache user requests during server idle time, reducing stress on the origin server. To reduce user access delays, web objects should be predicted and prefetched based on user use patterns, and then cached.
The majority of online pre-fetching research is based on historical user access patterns. If the history information suggests a high possibility of URL address A following B, B will be prefetched once A is accessed (Cheng-Zhong, 2000). 2
Web prefetching is the technique by which a proxy server obtains web pages in advance of a user’s request. When a client requests a web object, rather than making the request to the web server, it may be retrieved from the cache.
The primary consideration when choosing a web pre-fetching algorithm is its ability to forecast the web item to be prefetched in order to reduce latency. Web prefetching takes advantage of web pages’ spatial proximity
which means that pages connected to the current page are more likely to be viewed than other pages. Web prefetching can be used in a web context between clients and web servers, proxy servers and web servers, or clients and proxy servers (Greeshma, 2012).
Web prefetching approaches are divided into two categories: probability-based and clustering-based, which use weight functions. In probability-based pre-fetching, probabilities are derived based on data access history.
This approach assumes that the request sequence follows a pattern and calculates the likelihood of following that pattern. Clustering-based pre-fetching algorithms make decisions based on information from previously obtained web pages, assuming that pages nearby to previously fetched pages are more likely to be requested in the near future (Greeshma 2012).
Furthermore, web prefetching is a study issue that has received increased interest in recent years. The web pre-fetching gets some web assets before users request them.
Thus, cache pre-fetching aids in minimising user perceived latency. Many studies have demonstrated that combining caching and pre-fetching improves performance twice as much as single caching (Waleed, 2012). 3
Web caching is a well-known approach for increasing the performance of a Web-based system by storing Web items that will be needed in the near future in a location closer to the user.
Web caching mechanisms are implemented on three levels: client level, proxy level, and original server level. Significantly, proxy servers function as intermediaries between users and web sites, reducing response times and conserving network traffic. To improve response time, an efficient caching mechanism should be implemented in a proxy server (Waleed, 2011).
Due to limited cache space, an intelligent approach is required to efficiently handle Web cache content. Classical caching policies are inefficient in web caching because they examine only recency, frequency, and size while ignoring a combination of two elements that influence web caching efficiency.
Unfortunately, classical caching solutions do not significantly improve cache hit ratios. Even with an endless cache, the hit ratio remains limited regardless of the caching strategy.
This is because most individuals browse and investigate new web pages in an attempt to find fresh information. To overcome these constraints and boost cache hit ratio, the web pre-fetching technique is used with web caching.
Need help with a related project topic or New topic? Send Us Your Topic