Web Caching Research Articles

DYNASEM - An Improvised Dynamic and Semantic based Web Cache Replacement Policy

This paper proposes a web cache replacement policy based on semantic content of the pages cached at the client side. Two models namely Clustered Model(CM) and Relational Model(RM) are proposed that focus on the Dynamicity which refers the dynamic nature of the content and the Semantic content which exhibits the relation of information available among cached web pages and hence the name DynaSem. The proposed policy marks the page for eviction prioritized by Eviction Index (EI) in CM and Relation Index (RI) in RM. CM uses an interface with a web browser incorporated into it. The Trie data structure that enables the searching process to be more efficient has been framed to store the well-known categories of cached content as clusters. Pages with highest EI are marked for eviction. RM employs a technique to reveal the relation among cached documents. It evicts documents that are less related(minimum RI) to an incoming document which needs to be stored in the cache to ensure that only related documents are cached; hence the contents of the cache represent the documents of interest to the user and those which are of more static in nature. The proposed policy has been developed to incorporate two algorithms- one to find the dynamic count of the given web page 'P' and the other to the find semantic relation between the pages cached. Both the models(CM and RM) are used to establish the semantic relation. The policy has been simulated under model driven simulation with the help of an input set consisting of a few web pages. The parameters pertinent to cache replacement algorithms are computed and the result shows there is a factual improvement compared to the original semantic based policies. Keywords: Web caching, replacement policies, eviction, semantic relation, dynamism.

Configuration Management in Multi-Channel Multi-Container Web Application Servers

The Multi-Container Web server allows the division of a single web request into independent portions to be executed in parallel over different communication channels. To achieve this, the underlying communication infrastructure of traditional web environments is changed from the state-full TCP to the stateless UDP communication protocol. Such an architectural change provided an environment suitable for parallelization and distribution and enhanced some other already existing characteristics of the current web environments such as fault tolerance and web caching. In this paper, performance is further enhanced as part of the framework upon which web applications will run by introducing a number of changes. The configuration Manager (CM) is the subsystem that is responsible for reading configuration files provided by administrator. The CM is split into two parts, which are the Container CM and the HPA CM. After the CM reads the necessary configuration files and validates its content, it makes it easy for the administrators to be able to change environment setting such as container port ranges, types of nodes, services names, services location, number of instances each service should be loaded, and more. The CM also have the capability to reconstruct some types of configuration files from its data structure buffers, as such buffers may change through other subsystems and such changes may need to be permanent.

Web Cache Design and Implementation for Efficient SNMP Monitoring towards Internet-Scale Network Management

In this paper, we propose an SNMP-aware web cache design that has two main objectives: (1) to avoid overload of network devices by SNMP requests, and (2) guaranteeing the monitoring time granularity of SNMP Object Identifiers (OID) for a large scale network such as the Internet. To meet these objectives, a cache is built into an RESTful active proxy, called Tambourine, which is the gateway for accessing management information through the Internet. Tambourine changes the landscape of traditional SNMP monitoring by allowing the Internet users to monitor closed-domain network devices through translating requests in HTTP into SNMP. However, the typical web cache algorithm can not be used in Tambourine due to two main reasons: (1) SNMP is not a cache-aware protocol and therefore can not provide Tambourine with the caching rules that need to be applied, and (2) the cache in Tambourine needs to accommodate two SNMP monitoring patterns: periodic and on-demand polling. In order for efficient periodic polling, SNMP traffic is reduced by a multi-TTL cache and user (or Manager)-side aggregation. For efficient on-demand polling, four-state transition is used to categorize OIDs into dynamic and static objects, each of which is allocated an optimum TTL. To provide users with a proper time stamp, the cache time stamp is included in the response to the users' request. Our experiments show that our cache design gives the staleness of 0 and a bounded number of SNMP requests even when the number of users' requests goes to infinity.

A novel content classification scheme for web caches

Web caches are useful in reducing the user perceived latencies and web traffic congestion. Multi-level classification of web objects in caching is relatively an unexplored area. This paper proposes a novel classification scheme for web cache objects which utilizes a multinomial logistic regression (MLR) technique. The MLR model is trained to classify web objects using the information extracted from web logs. We introduce a novel grading parameter worthiness as a key for the object classification. Simulations are carried out with the datasets generated from real world trace files using the classifier in Least Recently Used-Class Based (LRU-C) and Least Recently Used-Multilevel Classes (LRU-M) cache models. Test results confirm that the proposed model has good online learning and prediction capability and suggest that the proposed approach is applicable to adaptive caching.

Web Cache Optimization in Semantic Based Web Search Engine

the tremendous growth of information available to end users through the Web, search engines come to play ever a more critical role. Nevertheless, because of their general-purpose approach, it is always less uncommon that obtained result sets provide a burden of useless pages. The next-generation Web architecture, represented by the Semantic Web, provides the layered architecture possibly allowing overcoming this limitation. The ontology for multiple search engines is written such that in this search engine for single query the final result is got from multiple search engines. After getting the user query result we can use the clustering. In this clustering the user query results is formed in the a to z form, the several search engines have been proposed, which allow increasing information retrieval accuracy by exploiting a key content of Semantic Web resources, that is, relations. We can use web cache optimization in search engine to get fast retrieval of user query results. In this work I have used web cache optimization based on eviction method for semantic web search engine. In this paper, analization of both advantages and disadvantages of some current Web cache replacement algorithms including lowest relative value algorithm, least weighted usage algorithm and least unified-value algorithm is done. Based on our analysis, we proposed a new algorithm, called least grade replacement (LGR), which takes recency, frequency, perfect- history, and document size into account for Web cache optimization.

Improving web cache server performance through arbitral thread and delayed caching

As the number of Internet users increase explosively, the delay in network response time is also increasing. An economic and efficient solution for this problem is web caching. But the use of a cache server can cause another bottleneck because of the concentration of requests at the cache server. Many studies on improving cache server performance have been suggested, but existing studies have focused on load balancing and/or caching capacity, not directly on improving the throughput of a single cache server. In this paper, we analyze the causes of cache server bottleneck, and propose an arbitral thread and the delayed caching mechanism as a solution. We use an arbitral thread in order to provide a quick service to users' service requests, and we use delayed caching in order to improve system reliability. The proposed cache server is implemented through a modification of the SQUID cache server, and we compare its performance with the original SQUID cache server.

A neural network proxy cache replacement strategy and its implementation in the Squid proxy server

As the Internet has become a more central aspect for information technology, so have concerns with supplying enough bandwidth and serving web requests to end users in an appropriate time frame. Web caching was introduced in the 1990s to help decrease network traffic, lessen user perceived lag, and reduce loads on origin servers by storing copies of web objects on servers closer to end users as opposed to forwarding all requests to the origin servers. Since web caches have limited space, web caches must effectively decide which objects are worth caching or replacing for other objects. This problem is known as cache replacement. We used neural networks to solve this problem and proposed the Neural Network Proxy Cache Replacement (NNPCR) method. The goal of this research is to implement NNPCR in a real environment like Squid proxy server. In order to do so, we propose an improved strategy of NNPCR referred to as NNPCR-2. We show how the improved model can be trained with up to twelve times more data and gain a 5–10% increase in Correct Classification Ratio (CCR) than NNPCR. We implemented NNPCR-2 in Squid proxy server and compared it with four other cache replacement strategies. In this paper, we use 84 times more data than NNPCR was tested against and present exhaustive test results for NNPCR-2 with different trace files and neural network structures. Our results demonstrate that NNPCR-2 made important, balanced decisions in relation to the hit rate and byte hit rate; the two performance metrics most commonly used to measure the performance of web proxy caches.

A Dynamic Web Caching Technique for using 'URL Rewriting'

In order to make surfing the internet faster, and to save redundant processing load with each request for the same web page, many caching techniques have been developed to reduce latency of retrieving data on World Wide Web. In this paper we will give a quick overview of existing web caching techniques used for dynamic web pages then we will introduce a design and implementation model that take advantage of “URL Rewriting” feature in some popular web servers, e.g. Apache, to provide an effective approach of caching dynamic web pages.

A novel server-side proxy caching strategy for large-scale multimedia applications

Nowadays, server-side Web caching becomes an important technique used to reduce the User Perceived Latency (UPL). In large-scale multimedia systems, there are many Web proxies, connected with a multimedia server, that can cache some most popular multimedia objects and respond to the requests for them. Multimedia objects have some particular characteristic, e.g., strict QoS requirements. Hence, even some efficient conventional caching strategies based on cache hit ratio, meant for non-multimedia objects, will confront some problems in dealing with the multimedia objects. If we consider additional resources of proxy besides cache space, say bandwidth, we can readily observe that high hit ratios may deteriorate the entire system performance. In this paper, we propose a novel placement model for networked multimedia systems, referred to as the H k / T model, which considers the combined influence of arrival rate, size, and playback time to select the objects to be cached. Based on this model, we propose an innovative Web caching algorithm, named as the ART-Greedy algorithm, which can balance the load among the proxies and achieve a minimum Average Response Time (ART) of the requests. Our experimental results conclusively demonstrate that the ART-Greedy algorithm outperforms the most popular and commonly used LFU (Least Frequently Used) algorithm significantly, and can achieve a better performance than the byte-hit algorithm when the system utilization is medium and high.

A Content-Oriented Web Cache Policy under P2P Video Distribution Systems

The increasing popularity of network-based multimedia applications poses many challenges for multimedia content providers to supply efficient and scalable multimedia services. Peer-to-Peer (P2P) networks have been shown to be a promising approach to provide large scale video services over Internet since they increase the scalability and robustness. In this paper, we propose and analyze an object management policy approach for video web cache in a P2P context, taking advantage of object's meta-data, e.g video popularity, and object's encoding techniques, e.g Scalable Video Coding (SVC). We exploit the peer-to-peer based approach to carry out the replication then eviction of videos in different web cache according to their popularity. Since videos are typically large and require high communication bandwidth for delivery, many peers may be unwilling to cache them in whole to serve others. Accordingly, we address this problem by proposing a new web caching approach for videos encoded in layers. Furthermore, in the proposed approach cooperation amongst the web caches may make it possible to replicate the video on several caches when this one is very popular. Afterward, we can gradually eliminate the replicas from the caches as the video popularity decrease. In addition, we structure our peer-to- peer network using interest-based communities. This network is made by ordinary nodes and super-nodes, and with peers clustered into communities. These communities are built around subjects (e.g., Science and Sports). The performance measurements we carried out allowed analyzing the traffic volume that flows from the community super-node to the first server into our content delivery network. The main goal is to measure how much our content-oriented management policy can help to decrease the demands put over the community output link. These measurements showed that the proposed approach (i) reduces the traffic volume, (ii) maximizes the overall bit rate of community demands, and (iii) optimizes the use of network resources.

CMS offline conditions framework and services

Non-event data describing detector conditions change with time and come from different data sources. They are accessible by physicists within the offline event-processing applications for precise calibration of reconstructed data as well as for data-quality control purposes. Over the past years CMS has developed and deployed a software system managing such data. Object-relational mapping and the relational abstraction layer of the LHC persistency framework are the foundation; the offline condition framework updates and delivers C++ data objects according to their validity. A high-level tag versioning system allows production managers to organize data in hierarchical view. A scripting API in python, command-line tools and a web service serve physicists in daily work. A mini-framework is available for handling data coming from external sources. Efficient data distribution over the worldwide network is guaranteed by a system of hierarchical web caches. The system has been tested and used in all major productions, test-beams and cosmic runs.

Performance impact of large file transfer on web proxy caching: A case study in a high bandwidth campus network environment

Since large objects consume substantial resources, web proxy caching incurs a fundamental trade-off between performance (i.e., hit-ratio and latency) and overhead (i.e., resource usage), in terms of caching and relaying large objects to users. This paper investigates how and to what extent the current dedicated-server based web proxy caching scheme is affected by large file transfers in a high bandwidth campus network environment. We use a series of trace-based performance analyses and profiling of various resource components in our experimental squid proxy cache server. Large file transfers often overwhelm our cache server. This causes a bottleneck in a web network, by saturating the network bandwidth of the cache server. Due to the requests for large objects, response times required for delivery of concurrently requested small objects increase, by a factor as high as a few million, in the worst cases. We argue that this cache bandwidth bottleneck problem is due to the fundamental limitations of the current centralized web proxy caching model that scales poorly when there are a limited amount of dedicated resources. This is a serious threat to the viability of the current web proxy caching model, particularly in a high bandwidth access network, since it leads to sporadic disconnections of the downstream access network from the global web network. We propose a peer-to-peer cooperative web caching scheme to address the cache bandwidth bottleneck problem. We show that it performs the task of caching and delivery of large objects in an efficient and cost-effective manner, without generating significant overheads for participating peers.

Alternative web caching design: a site-based approach

The existing designs and solutions for web caching systems commonly make caching decisions based on document or uniform resource locator (URL) information. While this conventional approach works, the authors deliver insights into the alternative approach using site information for web caching design. This site-based approach makes caching decisions based on the website that an object belongs to, rather than the object itself. The authors show that this new approach can benefit different scopes of cache design, ranging from internal operation of a single proxy (host level), mapping of proxy array in a local area network (LAN level), to load reduction in the global cache hierarchy (wireless area network (WAN) level). Since disks are usually the performance bottleneck in a proxy, to overcome this, the authors design a site-based cache architecture that tries to store web objects belonging to the same site in nearby disk blocks. The simulation results show that it reduces disk access time by 21–50%, compared to the conventional URL-based cache architecture. Besides, in the LAN-level design, a site-based mapping scheme can be used to map all requests targeting the same website to the same proxy, resulting in up to 50% reduction in the total transmission control protocol connection overhead. On the other hand, in the WAN-level, upper-level proxies are usually overloaded. To solve it, site information is used in lower-level proxies to decide which requests to forward to the origin servers directly (instead of upper-level proxy). As a result, proxy load is reduced by 46–59% and a reduction of 9–25% in request compliance delay can be achieved. With the merits of these three levels, it is believed that the site-based approach would contribute to the future caching designs.

Improved Security and Response Time by Using XML Web Service and Caching SSL Session Key

Improved Security and Response Time by Using XML Web Service and Caching SSL Session Key

Caching personalised and database-related dynamic web pages

In recent years, web development is the most important application in internet. Caching related technique improves the web server performance significantly. However, existing caching schemes cannot deal with the dynamic web pages efficiently. Thus, in this paper, we propose a caching scheme and then use web session objects and database-related dynamic web cache to implement the dynamic web cache system in Tomcat web server. We show how to build the dependency between dynamic web pages and the underlying database fields and session objects. Our experimental results show that Tomcat with proposed dynamic web cache can increase the stability of web server and improve web server throughput by up to 290%.

A Design and Implementation Model for Web Caching Using Server "URL Rewriting"

Abstract — In order to make surfing the internet faster, and to save redundant processing load with each request for the same web page, many caching techniques have been developed to reduce latency of retrieving data on World Wide Web. In this paper we will give a quick overview of existing web caching techniques used for dynamic web pages then we will introduce a design and implementation model that take advantage of “URL Rewriting” feature in some popular web servers, e.g. Apache, to provide an effective approach of caching dynamic web pages. Keywords — Web Caching, URL Rewriting, Optimizing Web Performance, Dynamic Web Pages Loading Time. I. I NTRODUCTION BOUT two decades ago, most of web pages on the internet were just static HTML pages and the job of building a website was an easy job and doesn’t require much knowledge of computer programming. But nowadays most of websites are dynamic ones which rely on a scripting language to run, and become more complicated as they provide new features such as personalization of contents and dynamic advertisements. Also it allows just about anyone, with limited or no web design experience, to update their own website via an administrative backend [7]. But unfortunately loading of a web dynamic page is slower and requires more operations than loading a static page [14], [15]. As a result the need of web caching becomes crucial as dynamic pages are the dominant in the web. To alleviate server processing overhead and reduce loading time of dynamic web page, it is possible to get the generated output of the dynamic page and save it in the cache, so that any subsequent request will be served from the cache only with no need to execute the same code again [12], [13]. Thus a dynamic web page can be served nearly as a static one. In this paper we will discuss the popular techniques of web caching used today, and then we will introduce a simple de-sign model and its implementation of a web caching technique followed by a comparison study and an analysis of the enhanced performance after using this technique.

Cryptographically Secure Bloom-Filters

In this paper, we propose a privacy-preserving variant of Bloom-filters. The Bloom-filter has many applications such as hash-based IP-traceback systems and Web cache sharing. In some of those appli...

Performance optimization for web caching based on input modeling

In this paper, the adaptive replacement policy and the pre-fetching technique for Web caching systems are studied with two user request models and their combinations. Performance results from two most commonly used replacement policies: LRU (Least Recently Used) and LFU (Least Frequently Used) are compared. It is shown that LRU performs better for one of the user request models and LFU is on the opposite. The adaptive replacement policy is intelligent in finding the best candidate policy as the user access pattern changes. The learning speed of the system is tested by different switch thresholds.

An analytical model for the performance evaluation of stack‐based Web cache replacement algorithms

AbstractWeb caching has been the solution of choice to web latency problems. The efficiency of a Web cache is strongly affected by the replacement algorithm used to decide which objects to evict once the cache is saturated. Numerous web cache replacement algorithms have appeared in the literature. Despite their diversity, a large number of them belong to a class known as stack‐based algorithms. These algorithms are evaluated mainly via trace‐driven simulation. The very few analytical models reported in the literature were targeted at one particular replacement algorithm, namely least recently used (LRU) or least frequently used (LFU). Further they provide a formula for the evaluation of the Hit Ratio only. The main contribution of this paper is an analytical model for the performance evaluation of any stack‐based web cache replacement algorithm. The model provides formulae for the prediction of the object Hit Ratio, the byte Hit Ratio, and the delay saving ratio. The model is validated against extensive discrete event trace‐driven simulations of the three popular stack‐based algorithms, LRU, LFU, and SIZE, using NLANR and DEC traces. Results show that the analytical model achieves very good accuracy. The mean error deviation between analytical and simulation results is at most 6% for LRU, 6% for the LFU, and 10% for the SIZE stack‐based algorithms. Copyright © 2009 John Wiley & Sons, Ltd.

Analyzing the Long Range Dependence and Object Popularity in Evaluating the Performance of Web Caching

Web cache systems enhance Web services by reducing the client side latency. To deploy an effective Web cache, analysis of the traffic characteristics is indispensable. Various reported results of traffic analysis show evidences of long range dependence (LRD) in the data stream and rank distribution of the documents in Web traffic. This article analyzes Web cache traffic properties like LRD and rank distribution based on the traces collected from NLANR (National Laboratory of Applied Network Research) cache servers. Traces are processed to investigate the performance of Web cache servers and traffic patterns. Statistical tools are utilized to measure the strengths of the LRD and popularity. The Hurst parameter, which is a measure of the LRD, is estimated using various statistical methods. It is observed that the presence of LRD in the traffic is feeble and does not influence the Web cache performance.