Web prefetching performance metrics: A survey

Although Internet service providers and communications companies are continuously offering higher and higher bandwidths, users still complain about the high latency they perceive when downloading pages from the web. Therefore, latency can be considered as the main web performance metric from the user's point of view. Many studies have demonstrated that web prefetching can be an interesting technique to reduce such latency at the expense of slightly increasing the network traffic. In this context, this paper presents an empirical study to investigate the maximum benefits that web users can expect from prefetching techniques in the current web. Unlike previous theoretical studies, this work considers a realistic prefetching architecture using real traces. In this way, the influence of real implementation constraints are considered and analyzed. The results obtained show that web prefetching could improve page latency up to 52% in the prefetchable part of the studied traces.

show abstract

“…Other performance indexes measured in this work are: object latency savings, byte recall, and object recall [20].…”

Section: Performance Measuresmentioning

confidence: 99%

Key factors in web latency savings in an experimental prefetching system

et al. 2011

Self Cite

View full text Add to dashboard Cite

show abstract

“…For this reason, the time taken to download a page should be considered as the main metric to evaluate the user-perceived latency [1]. Pages are usually composed of an HTML acting as a container of an increasing number of embedded objects like images, javascript code, CSS, etc.…”

Section: Downloading Time Issuesmentioning

confidence: 99%

Speculative Validation of Web Objects for Further Reducing the User-Perceived Latency

et al. 2010

Self Cite

View full text Add to dashboard Cite

Abstract. Web caching techniques reduce user-perceived latency by serving the most popular web objects from an intermediate memory.In order to assure that reused objects are not stale, conditional requests are sent to the origin web servers before serving them. Most of the server responses to the conditional requests ratify that the object remains valid and, as a consequence, they do not include the object itself. Therefore, the object transfer time is completely saved when the object is still valid. However, the round-trip time (RTT) of these short responses cannot be saved. This time represents an important fraction of the response time in the current Internet scenario and makes conditional requests save less perceived latency than when they were proposed. This paper proposes an approach to reduce the amount of conditional requests needed to maintain web cache consistency, thus completely saving both mentioned times (transfer and RTT) taken by such requests. To this end, our system uses a speculative approach similar to the one used in web prefetching which pre-sends freshness labels instead of web objects. The proposed technique has been evaluated using current and representative web traces. Experimental results show that the proposal dramatically reduces up to 55% of both the user-perceived latency and the amount of requests that the server receives.

show abstract

“…The variant Object Traffic Increase (∆Tr ob ) measures this cost taking into account only the amount of objects. As it was demonstrated in [14], the metrics that concern the prediction are interrelated as Equation 1 shows:…”

Section: A Performance Indexesmentioning

confidence: 99%

DDG: An Efficient Prefetching Algorithm for Current Web Generation

Domènech

Gil

Sahuquillo

et al. 2006

2006 1st IEEE Workshop on Hot Topics in Web Systems and Technologies

Self Cite

View full text Add to dashboard Cite

Abstract-Web prefetching is one of the techniques proposed to reduce user's perceived latencies in the World Wide Web. The spatial locality shown by user's accesses makes it possible to predict future accesses based on the previous ones. A prefetching engine uses these predictions to prefetch the web objects before the user demands them. The existing prediction algorithms achieved an acceptable performance when they were proposed but the high increase in the amount of embedded objects per page has reduced their effectiveness in the current web. In this paper we show that most of the predictions made by the existing algorithms are useless to reduce the user's perceived latency because these algorithms do not take into account how current web pages are structured, i.e., an HTML object with several embedded objects. Thus, they predict the accesses to the embedded objects in an HTML after reading the HTML itself. For this reason, the prediction advance is not enough to prefetch the objects and therefore there is no latency reduction. As a result of a wide analysis of the behaviour of the most commonly used algorithms, in this paper we present the DDG algorithm that distinguishes between container objects (HTML) and embedded objects to create a new prediction model according to the structure of the current web. Results show that, for the same amount of extra requests to the server, DDG always outperforms the existing algorithms by reducing the perceived latency between 15% and 150% more without increasing the computing complexity.

show abstract

Web prefetching performance metrics: A survey

Cited by 37 publications

References 25 publications

Key factors in web latency savings in an experimental prefetching system

Key factors in web latency savings in an experimental prefetching system

Speculative Validation of Web Objects for Further Reducing the User-Perceived Latency

DDG: An Efficient Prefetching Algorithm for Current Web Generation

Contact Info

Product

Resources

About