دانلود رایگان مقاله انگلیسی یک رویکرد مبتنی بر تشابه برای ذخیره سازی وب یکپارچه و تکرار محتوی در CDN ها به همراه ترجمه فارسی
عنوان فارسی مقاله | یک رویکرد مبتنی بر تشابه برای ذخیره سازی وب یکپارچه و تکرار محتوی در CDN ها |
عنوان انگلیسی مقاله | A similarity based approach for integrated Web caching and content replication in CDNs |
رشته های مرتبط | مهندسی کامپیوتر و فناوری اطلاعات، مهندسی نرم افزار، اینترنت و شبکه های گسترده، مهندسی الگوریتم ها و محاسبات و شبکه های کامپیوتری |
فرمت مقالات رایگان |
مقالات انگلیسی و ترجمه های فارسی رایگان با فرمت PDF آماده دانلود رایگان میباشند همچنین ترجمه مقاله با فرمت ورد نیز قابل خریداری و دانلود میباشد |
کیفیت ترجمه | کیفیت ترجمه این مقاله متوسط میباشد |
نشریه | آی تریپل ای – IEEE |
مجله | دهمین نشست بین المللی مهندسی و کاربردی پایگاه داده – 10th International Database Engineering and Applications Symposium |
سال انتشار | 2014 |
کد محصول | F544 |
مقاله انگلیسی رایگان (PDF) |
دانلود رایگان مقاله انگلیسی |
ترجمه فارسی رایگان (PDF) |
دانلود رایگان ترجمه مقاله |
خرید ترجمه با فرمت ورد |
خرید ترجمه مقاله با فرمت ورد |
جستجوی ترجمه مقالات | جستجوی ترجمه مقالات |
فهرست مقاله: چکیده 1 مقدمه 2 هدف مقاله 3 یکپارچه سازی تکرار حافظه نهان مبتنی بر تشابه 3-1 فرمولاسیون مسئله 3-2 روش پیشنهادی 4 ارزیابی و ازمایش عملکرد 4-1 بستر آزمایشی شبیه سازی 4-2 پارامتر های عملکرد 4-3 سیاست های مدیریت محتوی 4-4 ازمایش 5 نتیجه گیری |
بخشی از ترجمه فارسی مقاله: 1- مقدمه |
بخشی از مقاله انگلیسی: 1. Introduction The current network technology is challenged by the increasing growth of Web usage. A massive wave of simultaneous clients, might decrease the performance or even halt a state-of-the-art web server, the so-called “bottleneck” phenomenon [5]. Therefore, clients may experience latency or even denial of service (DoS). Two solutions have been proposed: Web caching and prefetching mainly applied in Web proxy servers and content replication currently implemented in the CDNs. Web caching temporarily stores the recently requested objects in cache to become available for future requests [6]. Web caching may be combined with prefetching which detects meaningful object access patterns predicting future requests [7] . Therefore, objects may be transferred to the proxy server a priori. Content replication creates replicas of objects as close to the clients as possible. Its current implementation is the Content Distribution Network (CDN), a trusted high-speed overlay network which is used to deliver Web objects, static data and streaming multimedia content [9]. Using pointsof-presence, the so called surrogate servers, CDNs manage to serve data faster and in a more reliable way. An important advantage of CDNs is that origin servers are protected from flash crowd events [10]. A flash crowd event occurs when numerous users access a Web site simultaneously, such as the one occurred in September 11th 2001 when users flooded popular news sites. Web caching is characterized by a dynamic nature in terms of its content update opposite to the static nature of replicated content in CDNs. A scheme that implements both under the same technology could demonstrate performance improvement. Therefore, we consider a CDN that its surrogate servers act both as cooperative replicators and proxy caches. We propose a method that shares the storage capacity of each surrogate server of a CDN for Web caching and content replication. Our method, called Similarity Replication Caching (SRC), uses the heuristic criterion of placement similarity. This paper is organized as follows. In Section 2 we discuss the motivation of this paper and present the related work. Section 3 propose the Web caching and static content replication integration problem along with our developed method. In Section 4 the experimentation is presented. Finally, the conclusion of our work are given in Section 5. 2. Motivation and Paper’s Contribution The content delivery service has a crucial impact in terms of CDN pricing and performance. In this work we focus on the placement of object replicas which is proved in [4] to be NP-complete. In general, an efficient replica placement reduces the cost of replica’s placement and improves the overall network performance [4, 8]. However, replicas’ placement is forced to be static for a considerable amount of time due to replication and distribution cost. This leads to inefficient storage capacity usage since the surrogate servers’ cache after a period of time would contain unwanted objects. The key issue is to reduce the impact of the disadvan tages of static replication by enabling partially dynamic replication. Towards this direction, a greedy Hybrid algorithm that combines an LRU cache replacement policy with static content replication on a CDN was proposed in [1]. The experimental results showed that the performance of the Hybrid algorithm outperformed the Web caching and static content replication as separate implementations. Motivated by this work, we further study this approach by presenting an alternative method which would detect automatically the appropriate percentage that splits the available storage capacity of each surrogate server for Web caching and static content replication. The challenge for us is to achieve a delicate balance between replication and caching towards improving the CDN’s performance under flash crowd events. Our contributions can be summarized as follows: • We propose a CDN where the surrogate servers act both as proxy caches and static replicators under a cooperative environment. • Develop a placement similarity approach (the so called SRC) a non parametric method, for evaluating the level of integration of Web caching with content replication. • Provide experimentation showing that our method performs better than the Hybrid algorithm [1] during flash crowd events and typical CDN load. |