Integrating VVVVVV Caches and Search Engines*

Size: px
Start display at page:

Download "Integrating VVVVVV Caches and Search Engines*"

Transcription

1 Global Internet: Application and Technology Integrating VVVVVV Caches and Search Engines* W. Meira Jr. R. Fonseca M. Cesario N. Ziviani Department of Computer Science Universidade Federal de Minas Gerais Belo Horizonte - MG - Brazil {meira, rfonseca. magc, nivio}@dcc.ufmg,br Abstract In this paper we propose the concept of cache plugins, which are customized programs that run WWW cache servers and perform some of the search engine tasks. We describe a prototype implementation of cache plugin to answer client requests directed to a large search engine, using a nearby cache server to store static objects. Experimental results using actual logs show a significant improvement on the quality of service of the search engine, doubling its predictability, improving its availability by a factor of 24, and reducing both its response time by 8% and the network traffic by a factor of Introduction The rapid and uncontrolled growth of the WWW has increased the importance of search engines as a means of finding relevant information, documents and services. These engines - have become the actual gateways - through which start navigation, and a measure Of their popularity and massive use is the cost of advertising in their sites, starting from a few thousand dollars a month. Because of their widespread use and relevance, efficient and fast searches are a need. However, the search servers' performance depends not only on the number of requests, but also on their complexity. The amount of data returned varies greatly among different requests, imposing a variable load to the servers and the interconnection networks between these and the clients. Another recent trend that has been gaining popular- ity is the use of meta-search engines [l], which aim at minimizing the users' effort by simultaneously querying several search engines and combining the results in a single page. Howerver, this can worsen the problem of performance degradation, since it multiplies the traffic generated by a single user. Standard caching strategies [4, 51, employed to minimize both response time and network traffic, are not adequate to cache responses to queries to search engines, which are intrinsically dynamic. These strategies only provide support to static objects, i.e., those which do not change frequently. Moreover, the replication of pages containing advertising material is avoided by content providers, since they can no longer control banner replacement policies, nor can they account for individual accesses to their pages. Even more elaborated strategies proposed to cache dynamic content i did not perform satisfactorily, increasing the client response time by up to 400% [3]. In this paper we propose cache as a new strategy for handling requests for non-static documents. Cache plugins are small programs that execute on hosts that act as WWW cache servers, enabling the cache to store static information from which the non-static pages can be built. Furthermore, they can act as concentrators for page view accounting. The paper is organized as follows. Section 2 describes the cache plugin architecture. In Section 3 introduces the Miner Family of Web agents as a good candidate for being improved in terms of its quality of service Meira Jr.). future work. Global Telecommunications Conference - Globecom' /99/ IEEE 1763

2 2 Cache Plugin Architecture In this section we describe the cache plugin architecture in detail and an implementation using Squid [6]. Cache plugins are programs that run either on the cache server or on a nearby (connectionwise) machine. Plugins answer requests for dynamic objects (i.e., objects that change frequently) that are normally ignored by WWW caches, being forwarded to remote servers. A busy content provider or search engine may implement a cache plugin for its site and make it available so that it can be installed in cache servers close to users. Once operational, the plugin is able to perform several tasks on behalf of the represented server, such as page generation and page view accounting, lowering both the load on the server and on the network. In our prototype implementation, requests reach the plugin through the redirection capability provided by Squid. Redirectors are scripts that rewrite URLs from requests, as specified by the cache administrator. In order to install a plugin we simply need to enable a redirection clause that will cause the cache server to redirect requests originally targetting a represented server to the respective local plugin. Upon receiving the request, the plugin parses it and determines all static objects, such as banner images, that are required to build the response. These objects are then requested to the cache server, which treats them as ordinary requests: if a local copy exists in the cache, it is returned; otherwise, the object is fetched from the remote host, sent to the plugin, and cached locally for future requests. Once all the required objects are available, the plugin combines them appropriately, building the response page, which is sent to the cache server and finally to the client. Furthermore, the plugin can gather information, such as accounting of page accesses and clicks in advertising banners, and report it periodically to the server it represents. This process is completely transparent to the client, and is summarized in Figure 1. The numbers in the figure refer to the following seven steps: 1) client requests a non-static page from a cache server; 2) cache server redirects the request to the cache plugin; 3) cache plugin parses the request and request static objects from the cache server; 4) static objects may be misses and are requested from original servers; 5) plugin composes response page and answers cache server s request; 6) cache server returns page to client; and 7) cache plugin notifies server about accesses and may retrieve new procedures for response page generation. It is important to note that each content provider implements and distributes its own plugin, which is a I Client I Figure 1: General Plugin Architecture therefore tailored to understand requests directed to its represented server, being able to determine what data to request from the server, and how to interpret the data. This can involve the use of encryption to enforce security,for example. Moreover, step 7 in Figure 1, the plugin reporting back to the server, can involve any kind of communication, since the protocol is defined by the content provider. The freshness of the objects in the cache can be controlled by ordinary TTL (time to live) mechanisms. The server can set the TTL of a static object returned based on how often it is uptdated. The use of the cache plugins is beneficial for both the remote server, which experiences a reduction in processor and network load, and the client making the request, which has its response time lowered. Obviously, gains from employing cache plugins depend on the reference locality in the request stream to which the plugin is subject. 3 The Miner Family of Web Agents An example of an application that can benefit from the cache plugin architecture is the Miner Family of Web Agents [l]. The Miner Family is a collection of individual programs, called agents, that can act both as searching utilities and an electronic catalogs, and can also provide brokerage services. The Miner Family was developed mainly for Portuguese language-based ser Global Telecommunications Conference - Globecom 99

3 vices. The search utility services provided by the Miner Family include MetaMiner, which is a meta-search engine that uses Brazilian and international search engines, among others. The Miner Family was coded in Java and comprises about 23,000 lines of code that run on a Netscape Enterprise Server. All members of the Miner Family work similarly and the main steps to answer a request can be summarized as follows: (1) a user submits a query; (2) the Miner server gets the query and dispatches its agents; (3) each agent queries its target engine, store, or site; (4) each agent receives and parses the query results; and (5) the server unifies, formats, and sends the results to the user. The opportunity for caching lies in the fact that all agents, in some point, combine static objects to form a request. Each query to an agent of the Miner Family is specified by of one or more words and a few additional parameters. Using the Meta Miner agent as an example, if the server executing the agent holds lists of URLs for each of the words making up the query, building the response to the query would just require the agent to appropriately merge the lists together. These lists can also be sent to a remote plugin, cached and combined locally for future requests. 4 Experimental Results In order to verify the efficiency of the cache plugin architecture we evaluated the gains observed in terms of quality of service of the search engine, using a workload derived from MetaMiner logs. This workload is used to drive an experimental environment consisting of clients that submit requests to a cache server in which a cache plugin is running. This cache plugin represents a search engine server that runs in a separate machine. Both the cache plugin and the search engine server build the response to a query similarly: they merge the URL lists corresponding to the words contained in the query. Since these lists remain unchanged for a relatively long period, they are considered to be static here, and can be cached normally. To obtain the workload for our experiments we combined data from two different sources: (1) logs from the MetaMiner engine, which held a record of each request submitted, with all options the user chose at the moment of submitting it, and (2) text documents retrieved from the Web (those with Mime type text/*) that were stored in the disks of POP-MG, a large proxy cache server in Brazil. The MetaMiner logs span 2 br months and comprised a total of 925,042 requests to the engine. We noticed a high degree of reference locality among search terms, with the hit ratio for a period of 24 hours averaging 80%. The requests had an average of 1.63 words, with 52% consisting of only one word and the largest query being 12-word long. We processed the logs and isolated all the words that made up the queries, for a total of 125,342 unique words. We searched for each of these words in the database of text documents retrieved from the POP- MG cache server, and built a list containing all documents where that word appeared. We retrieved 85,426 documents from the cache, that comprise a total of megabytes of text data. The average size of the URL lists for each word is 38,90has 6967 URLs. By using this approach to generate the workload, we garanteed a realistic distribution in the size of the lists and in the frequency of requisition for each list, since both the popularity of a term in actual documents and in actual searches are taken into account. The experiments consisted of several clients submitting requests to the cache server. In each run, the requests are a subset of the requests present in the MetaMiner logs, and are formed by one or more words. Our experimental environment consists of Pentium machines running Solaris and FreeBSD in the same switched LAN. The machine that acts as a search engine server is a Pentium Pro SMP with four processors] 128 Mb of memory, and Ultra-wide SCSI disks. The cache server is a dual Pentium Pro SMP, 128 Mb of memory, and Ultra-wide SCSI disks. Finally, the clients are Pentium 166Mhz with 64Mb of memory. The clients are processes that emulate a set of browsers performing queries. Each client queries a single cache server, keeping a configurable number of requests open. The clients are able to handle efficiently several simultaneous connections by using asynchronous communication primitives, as proposed in [2], The server is a Perl script that parses the queries, retrieves the lists of URLs for each word that compose the query, joins these lists identifying the URLs that are present in all lists, and generates a page to be returned to the client. The cache redirector and the plugin are also Perl scripts. The redirector is quite simple and just translates server requests to requests to the plugin. The plugin parses the request and determines which lists of URLs should be requested to the cache server and later combined to generate the response to the client. The generated pages, in contrast to URL lists, are not cached, in order to optimize the cache disk utilization. We evaluated the efficacy of cache plugins through Global Telecommunications Conference - Globecorn

4 Globol Internet: Application and Technology Measures Average Response Time (Sec.) Response Time Relative Variance Errors Server CPU Load (usr/sys/wio) Server Disk Load Cache to Server Requests Bvtes Transferred /Sec. Plugin No Plugin /0.01/ /0.17/ Table 1: Profiles four metrics that quantify the quality of service of a search engine: response time, availability, predictability, and scalability. In order to evaluate these metrics, we performed two experiments that differ only by the use or not of cache plugins. There were two client processes per experiment; each client kept 20 connections open simultaneously and performed 80,000 requests from the workload described earlier in this section. The cache server keeps 20 instances of the cache plugin running. Table 1 shows some performance measures that we discuss next. We can observe that the use of cache plugins reduced the client response time by 8%. This relatively small improvement is explained by the prototypical implementation of plugins, which performs several forks, operations that are well known for being computationally expensive. Moreover, both response time variance and number of errors (i.e., requests that timed out without response) are much higher for the nonplugin c6nfiguration, indicating its lower predictability of service latency and service availability, respectively. In Figure 2 we show the average response time over the course of the experiments for one of the clients. One can observe the higher variance and the slightly higher average response time for the non-plugin configuration. Regarding server load, both CPU and disk loads decreased by one order of magnitude when employing cache plugins. The two last rows in the table present the gains in terms of the number of requests to the search engine server, which reduced by a factor of 10, and network traffic (measured by bytes transferred per second), which reduced by a factor of 20. Finally, we evaluated the service scalability by variating the number of simultaneous requests that are submitted by clients. We performed similar experiments with just 5 simultaneous requests per client and compared the response times variations. The average response time when we used the cache plugin was 1.35 seconds, compared to 1.82 seconds without the plugin, an improvement rate of 24%. The lower improvement rate pro- vided under heavier workloads is explained by the CPU saturation experienced by the cache server. Note that although the search engine server is twice as powerful as the cache server, we were still able to perform significantly better. We believe that the gains would be even higher in real environments, where there may be several cache servers representing a single search engine, and worse network conditions, making the access to the search engine more difficult. Recall that the gains just presented were a consequence of the reference locality present in the stream of requests, which can be further exploited by prefetching popular lists of URLs. Prefetching would allow not only faster response, but also congestion avoidance, since it can be performed during off-peak time periods. In order to evaluate the applicability of prefetching, we ran additional experiments in which each client warmed up the cache server with the 16,000 most popular URL lists from the first experiment, and then requested a different stream of 80,000 requests from the workload. The results are presented in the graph of Figure 2 ( Plugin with Prefetching ), where we can see that prefetching further improved the average response time by lo%, decreasing it to 1.22 seconds. 5 Conclusions and Future Work In this paper we presented Cache Plugins, a novel strategy to improve the quality of service of searches in the WWW, by integrating search engines and cache servers. Using the proposed architecture, result pages can be effectively cached, as well as any dynamic content page, normally not cacheable. This strategy reduces both the response time seen by client, and the server and network load. The implementation of such cache plugins is straightforward, and they also allow accounting of page visits and advertising clickthrough, one of the main restrictions for caching certain pages with dynamic content. We are to investigate the benefits of the architecture with data from a real search Global Telecommunications Conference - Globecom 99

5 Average Client Response Time I I I I 1 I I 1 j Plugin - Plugin withiprefetching No Plugin._.._... 0' I I I 1 I I I I Requests(Thousands) Figure 2: Average client response time engine, which has larger lists of URLs, and the implementation of more sophisticated protocols between the plugin and the server, so that it can decide whether to request the dynamic page directly or the static objects necessary to form the response. we also intend to investigate the use of cache plugins in other scenarios, such as electronic commerce. References [4] Calos Maltzahn, Kathy J. Richardson, and Dirk Grunwald. Performance Issues of Enterprise Level Web Proxies. ACM Sigmetn'cs '97, [5] C. Roadknight and I. Marshall. Variations in cache behavior. In Proceedings of WWW7, (61 Duane Wessels and K. Claffy. Squid Internet Object Cache, http : //www. nlanr. net/squid. [I] V. Almeida, W. Meira Jr., V. Ftibeiro, and N. Ziviani. Efficiency analysis of e-brokers in the' electronic marketplace. In Proceedings of WWWB, [2] Gaurav Banga and Peter Druschel. Measuring the Capacity of a Web Server. In Usenix Symposium on Internet Technologies and Systems, Monterey, December [3] P. Cao, J. Zhang, and Kevin Beach. Active cache: Caching dynamic contents on the web. In Proc. of IFIP International Conference on Distributed Systems Platforms and Open Distributed Processing, pages ,1998. Global Telecommunications Conference - Globecom'

E-representative: a scalability scheme for e-commerce

E-representative: a scalability scheme for e-commerce E-representative: a scalability scheme for e-commerce Wagner Meira Jr. y Daniel Menascé z Virgílio Almeida y Rodrigo Fonseca y y Dept. of Computer Science z Dept. of Computer Science Universidade Federal

More information

A Scalable Approach for the Distribution of E-commerce Services Based on Application Level Active Networks

A Scalable Approach for the Distribution of E-commerce Services Based on Application Level Active Networks A Scalable Approach for the Distribution of E-commerce Services Based on Application Level Active Networks Fabrício Benevenuto, Breno Vitorino, Bruno Coutinho, Dorgival Guedes, Wagner Meira Jr. 1 Departamento

More information

Evaluating Cache-Layering to Improve Web Cache System Performance

Evaluating Cache-Layering to Improve Web Cache System Performance Evaluating Cache-Layering to Improve Web Cache System Performance Bruno Abrahão Fabrício Benevenuto Departamento de Ciência da Computação Universidade Federal de Minas Gerais Av. Antônio Carlos, 6627,

More information

Evaluation of Performance of Cooperative Web Caching with Web Polygraph

Evaluation of Performance of Cooperative Web Caching with Web Polygraph Evaluation of Performance of Cooperative Web Caching with Web Polygraph Ping Du Jaspal Subhlok Department of Computer Science University of Houston Houston, TX 77204 {pdu, jaspal}@uh.edu Abstract This

More information

Design and Implementation of A P2P Cooperative Proxy Cache System

Design and Implementation of A P2P Cooperative Proxy Cache System Design and Implementation of A PP Cooperative Proxy Cache System James Z. Wang Vipul Bhulawala Department of Computer Science Clemson University, Box 40974 Clemson, SC 94-0974, USA +1-84--778 {jzwang,

More information

Characterizing Home Pages 1

Characterizing Home Pages 1 Characterizing Home Pages 1 Xubin He and Qing Yang Dept. of Electrical and Computer Engineering University of Rhode Island Kingston, RI 881, USA Abstract Home pages are very important for any successful

More information

On the Interaction Between an Operating System and Web Server

On the Interaction Between an Operating System and Web Server On the Interaction Between an Operating System and Web Server David J. Yates (djy@cs.bu.edu) Computer Science Department Boston University Boston, MA 2215 USA Virgílio Almeida (virgilio@dcc.ufmg.br) Jussara

More information

Multimedia Streaming. Mike Zink

Multimedia Streaming. Mike Zink Multimedia Streaming Mike Zink Technical Challenges Servers (and proxy caches) storage continuous media streams, e.g.: 4000 movies * 90 minutes * 10 Mbps (DVD) = 27.0 TB 15 Mbps = 40.5 TB 36 Mbps (BluRay)=

More information

Modelling and Analysis of Push Caching

Modelling and Analysis of Push Caching Modelling and Analysis of Push Caching R. G. DE SILVA School of Information Systems, Technology & Management University of New South Wales Sydney 2052 AUSTRALIA Abstract: - In e-commerce applications,

More information

A CONTENT-TYPE BASED EVALUATION OF WEB CACHE REPLACEMENT POLICIES

A CONTENT-TYPE BASED EVALUATION OF WEB CACHE REPLACEMENT POLICIES A CONTENT-TYPE BASED EVALUATION OF WEB CACHE REPLACEMENT POLICIES F.J. González-Cañete, E. Casilari, A. Triviño-Cabrera Department of Electronic Technology, University of Málaga, Spain University of Málaga,

More information

because all web objects are not cacheable [3]. Data compression algorithms are applied on the information content of the web object.

because all web objects are not cacheable [3]. Data compression algorithms are applied on the information content of the web object. Development of an Optimization Algorithm for Internet Data Traffic Syed Misbahuddin and Tariq Ibn Aziz Department of Computer Science and Software Engineering Hail Community College King Fahd University

More information

On the Relationship of Server Disk Workloads and Client File Requests

On the Relationship of Server Disk Workloads and Client File Requests On the Relationship of Server Workloads and Client File Requests John R. Heath Department of Computer Science University of Southern Maine Portland, Maine 43 Stephen A.R. Houser University Computing Technologies

More information

A NEW PERFORMANCE EVALUATION TECHNIQUE FOR WEB INFORMATION RETRIEVAL SYSTEMS

A NEW PERFORMANCE EVALUATION TECHNIQUE FOR WEB INFORMATION RETRIEVAL SYSTEMS A NEW PERFORMANCE EVALUATION TECHNIQUE FOR WEB INFORMATION RETRIEVAL SYSTEMS Fidel Cacheda, Francisco Puentes, Victor Carneiro Department of Information and Communications Technologies, University of A

More information

Resource Placement in Distributed e-commerce Servers

Resource Placement in Distributed e-commerce Servers Resource Placement in Distributed e-commerce Servers Gustavo M. C. Gama Wagner Meira Jr. Márcio L. B. Carvalho Dorgival O. Guedes Virgílio A. F. Almeida e-speed Laboratory - Computer Science Department

More information

Seminar on. By Sai Rahul Reddy P. 2/2/2005 Web Caching 1

Seminar on. By Sai Rahul Reddy P. 2/2/2005 Web Caching 1 Seminar on By Sai Rahul Reddy P 2/2/2005 Web Caching 1 Topics covered 1. Why Caching 2. Advantages of Caching 3. Disadvantages of Caching 4. Cache-Control HTTP Headers 5. Proxy Caching 6. Caching architectures

More information

Summary Cache based Co-operative Proxies

Summary Cache based Co-operative Proxies Summary Cache based Co-operative Proxies Project No: 1 Group No: 21 Vijay Gabale (07305004) Sagar Bijwe (07305023) 12 th November, 2007 1 Abstract Summary Cache based proxies cooperate behind a bottleneck

More information

Edge Side Includes (ESI) Overview

Edge Side Includes (ESI) Overview Edge Side Includes (ESI) Overview Abstract: Edge Side Includes (ESI) accelerates dynamic Web-based applications by defining a simple markup language to describe cacheable and non-cacheable Web page components

More information

CHAPTER 4 OPTIMIZATION OF WEB CACHING PERFORMANCE BY CLUSTERING-BASED PRE-FETCHING TECHNIQUE USING MODIFIED ART1 (MART1)

CHAPTER 4 OPTIMIZATION OF WEB CACHING PERFORMANCE BY CLUSTERING-BASED PRE-FETCHING TECHNIQUE USING MODIFIED ART1 (MART1) 71 CHAPTER 4 OPTIMIZATION OF WEB CACHING PERFORMANCE BY CLUSTERING-BASED PRE-FETCHING TECHNIQUE USING MODIFIED ART1 (MART1) 4.1 INTRODUCTION One of the prime research objectives of this thesis is to optimize

More information

Correlation based File Prefetching Approach for Hadoop

Correlation based File Prefetching Approach for Hadoop IEEE 2nd International Conference on Cloud Computing Technology and Science Correlation based File Prefetching Approach for Hadoop Bo Dong 1, Xiao Zhong 2, Qinghua Zheng 1, Lirong Jian 2, Jian Liu 1, Jie

More information

AccWeb Improving Web Performance via Prefetching

AccWeb Improving Web Performance via Prefetching AccWeb Improving Web Performance via Prefetching Qizhe Cai Wei Hu Yueyang Qiu {qizhec,huwei,yqiu}@cs.princeton.edu Abstract We present AccWeb (Accelerated Web), a web service that improves user experience

More information

HTRC Data API Performance Study

HTRC Data API Performance Study HTRC Data API Performance Study Yiming Sun, Beth Plale, Jiaan Zeng Amazon Indiana University Bloomington {plale, jiaazeng}@cs.indiana.edu Abstract HathiTrust Research Center (HTRC) allows users to access

More information

Computer Networks. Wenzhong Li. Nanjing University

Computer Networks. Wenzhong Li. Nanjing University Computer Networks Wenzhong Li Nanjing University 1 Chapter 8. Internet Applications Internet Applications Overview Domain Name Service (DNS) Electronic Mail File Transfer Protocol (FTP) WWW and HTTP Content

More information

Enhanced Performance of Database by Automated Self-Tuned Systems

Enhanced Performance of Database by Automated Self-Tuned Systems 22 Enhanced Performance of Database by Automated Self-Tuned Systems Ankit Verma Department of Computer Science & Engineering, I.T.M. University, Gurgaon (122017) ankit.verma.aquarius@gmail.com Abstract

More information

Maintaining Mutual Consistency for Cached Web Objects

Maintaining Mutual Consistency for Cached Web Objects Maintaining Mutual Consistency for Cached Web Objects Bhuvan Urgaonkar, Anoop George Ninan, Mohammad Salimullah Raunak Prashant Shenoy and Krithi Ramamritham Department of Computer Science, University

More information

Web-based Energy-efficient Cache Invalidation in Wireless Mobile Environment

Web-based Energy-efficient Cache Invalidation in Wireless Mobile Environment Web-based Energy-efficient Cache Invalidation in Wireless Mobile Environment Y.-K. Chang, M.-H. Hong, and Y.-W. Ting Dept. of Computer Science & Information Engineering, National Cheng Kung University

More information

Threshold-Based Markov Prefetchers

Threshold-Based Markov Prefetchers Threshold-Based Markov Prefetchers Carlos Marchani Tamer Mohamed Lerzan Celikkanat George AbiNader Rice University, Department of Electrical and Computer Engineering ELEC 525, Spring 26 Abstract In this

More information

Chapter The LRU* WWW proxy cache document replacement algorithm

Chapter The LRU* WWW proxy cache document replacement algorithm Chapter The LRU* WWW proxy cache document replacement algorithm Chung-yi Chang, The Waikato Polytechnic, Hamilton, New Zealand, itjlc@twp.ac.nz Tony McGregor, University of Waikato, Hamilton, New Zealand,

More information

Trace Driven Simulation of GDSF# and Existing Caching Algorithms for Web Proxy Servers

Trace Driven Simulation of GDSF# and Existing Caching Algorithms for Web Proxy Servers Proceeding of the 9th WSEAS Int. Conference on Data Networks, Communications, Computers, Trinidad and Tobago, November 5-7, 2007 378 Trace Driven Simulation of GDSF# and Existing Caching Algorithms for

More information

Improve Web Application Performance with Zend Platform

Improve Web Application Performance with Zend Platform Improve Web Application Performance with Zend Platform Shahar Evron Zend Sr. PHP Specialist Copyright 2007, Zend Technologies Inc. Agenda Benchmark Setup Comprehensive Performance Multilayered Caching

More information

A New Architecture for HTTP Proxies Using Workstation Caches

A New Architecture for HTTP Proxies Using Workstation Caches A New Architecture for HTTP Proxies Using Workstation Caches Author Names : Prabhakar T V (tvprabs@cedt.iisc.ernet.in) R. Venkatesha Prasad (vprasad@cedt.iisc.ernet.in) Kartik M (kartik_m@msn.com) Kiran

More information

Netegrity SiteMinder 4.51 AuthMark Performance Details

Netegrity SiteMinder 4.51 AuthMark Performance Details Page 1 of 12 Netegrity SiteMinder 4.51 AuthMark Performance Details By Bruce Weiner (PDF version, 96 KB) Contents Executive Summary Test Methodology iload MVP AuthMark Result Analysis Server Hardware Server

More information

A Simulation: Improving Throughput and Reducing PCI Bus Traffic by. Caching Server Requests using a Network Processor with Memory

A Simulation: Improving Throughput and Reducing PCI Bus Traffic by. Caching Server Requests using a Network Processor with Memory Shawn Koch Mark Doughty ELEC 525 4/23/02 A Simulation: Improving Throughput and Reducing PCI Bus Traffic by Caching Server Requests using a Network Processor with Memory 1 Motivation and Concept The goal

More information

Architecture Tuning Study: the SimpleScalar Experience

Architecture Tuning Study: the SimpleScalar Experience Architecture Tuning Study: the SimpleScalar Experience Jianfeng Yang Yiqun Cao December 5, 2005 Abstract SimpleScalar is software toolset designed for modeling and simulation of processor performance.

More information

Improving object cache performance through selective placement

Improving object cache performance through selective placement University of Wollongong Research Online Faculty of Informatics - Papers (Archive) Faculty of Engineering and Information Sciences 2006 Improving object cache performance through selective placement Saied

More information

Evaluating external network bandwidth load for Google Apps

Evaluating external network bandwidth load for Google Apps Evaluating external network bandwidth load for Google Apps This document describes how to perform measurements to better understand how much network load will be caused by using a software as a service

More information

Finding a needle in Haystack: Facebook's photo storage

Finding a needle in Haystack: Facebook's photo storage Finding a needle in Haystack: Facebook's photo storage The paper is written at facebook and describes a object storage system called Haystack. Since facebook processes a lot of photos (20 petabytes total,

More information

Performance Modeling and Evaluation of Web Systems with Proxy Caching

Performance Modeling and Evaluation of Web Systems with Proxy Caching Performance Modeling and Evaluation of Web Systems with Proxy Caching Yasuyuki FUJITA, Masayuki MURATA and Hideo MIYAHARA a a Department of Infomatics and Mathematical Science Graduate School of Engineering

More information

Surveying Formal and Practical Approaches for Optimal Placement of Replicas on the Web

Surveying Formal and Practical Approaches for Optimal Placement of Replicas on the Web Surveying Formal and Practical Approaches for Optimal Placement of Replicas on the Web TR020701 April 2002 Erbil Yilmaz Department of Computer Science The Florida State University Tallahassee, FL 32306

More information

THE EFFECT OF JOIN SELECTIVITIES ON OPTIMAL NESTING ORDER

THE EFFECT OF JOIN SELECTIVITIES ON OPTIMAL NESTING ORDER THE EFFECT OF JOIN SELECTIVITIES ON OPTIMAL NESTING ORDER Akhil Kumar and Michael Stonebraker EECS Department University of California Berkeley, Ca., 94720 Abstract A heuristic query optimizer must choose

More information

Microsoft SQL Server Fix Pack 15. Reference IBM

Microsoft SQL Server Fix Pack 15. Reference IBM Microsoft SQL Server 6.3.1 Fix Pack 15 Reference IBM Microsoft SQL Server 6.3.1 Fix Pack 15 Reference IBM Note Before using this information and the product it supports, read the information in Notices

More information

Subway : Peer-To-Peer Clustering of Clients for Web Proxy

Subway : Peer-To-Peer Clustering of Clients for Web Proxy Subway : Peer-To-Peer Clustering of Clients for Web Proxy Kyungbaek Kim and Daeyeon Park Department of Electrical Engineering & Computer Science, Division of Electrical Engineering, Korea Advanced Institute

More information

Design and Performance Evaluation of Networked Storage Architectures

Design and Performance Evaluation of Networked Storage Architectures Design and Performance Evaluation of Networked Storage Architectures Xubin He (Hexb@ele.uri.edu) July 25,2002 Dept. of Electrical and Computer Engineering University of Rhode Island Outline Introduction

More information

INTRODUCTION. Chapter GENERAL

INTRODUCTION. Chapter GENERAL Chapter 1 INTRODUCTION 1.1 GENERAL The World Wide Web (WWW) [1] is a system of interlinked hypertext documents accessed via the Internet. It is an interactive world of shared information through which

More information

A Quantitative Analysis of the User Behavior of a Large E-Broker æ

A Quantitative Analysis of the User Behavior of a Large E-Broker æ A Quantitative Analysis of the User Behavior of a Large E-Broker æ Virgílio Almeida y Wagner Meira Jr. y Victor Ribeiro z Nivio Ziviani yz y Dept. of Computer Science z Miner Technology Group Univ. Federal

More information

Measurement-based Analysis of TCP/IP Processing Requirements

Measurement-based Analysis of TCP/IP Processing Requirements Measurement-based Analysis of TCP/IP Processing Requirements Srihari Makineni Ravi Iyer Communications Technology Lab Intel Corporation {srihari.makineni, ravishankar.iyer}@intel.com Abstract With the

More information

WebSphere Application Server Base Performance

WebSphere Application Server Base Performance WebSphere Application Server Base Performance ii WebSphere Application Server Base Performance Contents WebSphere Application Server Base Performance............. 1 Introduction to the WebSphere Application

More information

A Comparative Study of Web Prefetching Techniques Focusing on User s Perspective

A Comparative Study of Web Prefetching Techniques Focusing on User s Perspective A Comparative Study of Web Prefetching Techniques Focusing on User s Perspective Josep Domènech Ana Pont Julio Sahuquillo José A. Gil Department of Computing Engineering (DISCA) Universitat Politècnica

More information

Configuring Caching Services

Configuring Caching Services CHAPTER 8 This chapter describes how to configure conventional caching services (HTTP, FTP [FTP-over-HTTP caching and native FTP caching], HTTPS, and DNS caching) for centrally managed Content Engines.

More information

Power and Locality Aware Request Distribution Technical Report Heungki Lee, Gopinath Vageesan and Eun Jung Kim Texas A&M University College Station

Power and Locality Aware Request Distribution Technical Report Heungki Lee, Gopinath Vageesan and Eun Jung Kim Texas A&M University College Station Power and Locality Aware Request Distribution Technical Report Heungki Lee, Gopinath Vageesan and Eun Jung Kim Texas A&M University College Station Abstract With the growing use of cluster systems in file

More information

Analyzing Imbalance among Homogeneous Index Servers in a Web Search System

Analyzing Imbalance among Homogeneous Index Servers in a Web Search System Analyzing Imbalance among Homogeneous Index Servers in a Web Search System C.S. Badue a,, R. Baeza-Yates b, B. Ribeiro-Neto a,c, A. Ziviani d, N. Ziviani a a Department of Computer Science, Federal University

More information

BEx Front end Performance

BEx Front end Performance BUSINESS INFORMATION WAREHOUSE BEx Front end Performance Performance Analyses of BEx Analyzer and Web Application in the Local and Wide Area Networks Environment Document Version 1.1 March 2002 Page 2

More information

Oracle Enterprise Manager. 1 Before You Install. System Monitoring Plug-in for Oracle Unified Directory User's Guide Release 1.0

Oracle Enterprise Manager. 1 Before You Install. System Monitoring Plug-in for Oracle Unified Directory User's Guide Release 1.0 Oracle Enterprise Manager System Monitoring Plug-in for Oracle Unified Directory User's Guide Release 1.0 E24476-01 October 2011 The System Monitoring Plug-In for Oracle Unified Directory extends Oracle

More information

A Comparison of File. D. Roselli, J. R. Lorch, T. E. Anderson Proc USENIX Annual Technical Conference

A Comparison of File. D. Roselli, J. R. Lorch, T. E. Anderson Proc USENIX Annual Technical Conference A Comparison of File System Workloads D. Roselli, J. R. Lorch, T. E. Anderson Proc. 2000 USENIX Annual Technical Conference File System Performance Integral component of overall system performance Optimised

More information

Modification and Evaluation of Linux I/O Schedulers

Modification and Evaluation of Linux I/O Schedulers Modification and Evaluation of Linux I/O Schedulers 1 Asad Naweed, Joe Di Natale, and Sarah J Andrabi University of North Carolina at Chapel Hill Abstract In this paper we present three different Linux

More information

COMMON INTERNET FILE SYSTEM PROXY

COMMON INTERNET FILE SYSTEM PROXY COMMON INTERNET FILE SYSTEM PROXY CS739 PROJECT REPORT ANURAG GUPTA, DONGQIAO LI {anurag, dongqiao}@cs.wisc.edu Computer Sciences Department University of Wisconsin, Madison Madison 53706, WI May 15, 1999

More information

CIT 668: System Architecture. Caching

CIT 668: System Architecture. Caching CIT 668: System Architecture Caching Topics 1. Cache Types 2. Web Caching 3. Replacement Algorithms 4. Distributed Caches 5. memcached A cache is a system component that stores data so that future requests

More information

An Oracle White Paper. Released April 2013

An Oracle White Paper. Released April 2013 Performance and Scalability Benchmark: Siebel CRM Release 8.1.1.4 Industry Applications and Oracle 11.2.0.3 Database on Oracle's SPARC T5 Servers and Oracle Solaris An Oracle White Paper Released April

More information

Technical Brief: Specifying a PC for Mascot

Technical Brief: Specifying a PC for Mascot Technical Brief: Specifying a PC for Mascot Matrix Science 8 Wyndham Place London W1H 1PP United Kingdom Tel: +44 (0)20 7723 2142 Fax: +44 (0)20 7725 9360 info@matrixscience.com http://www.matrixscience.com

More information

SEDA: An Architecture for Well-Conditioned, Scalable Internet Services

SEDA: An Architecture for Well-Conditioned, Scalable Internet Services SEDA: An Architecture for Well-Conditioned, Scalable Internet Services Matt Welsh, David Culler, and Eric Brewer Computer Science Division University of California, Berkeley Operating Systems Principles

More information

CoBWeb A Crawler for the Brazilian Web

CoBWeb A Crawler for the Brazilian Web CoBWeb A Crawler for the Brazilian Web Altigran S. da Silva Eveline A. Veloso Paulo B. Golgher Berthier Ribeiro-Neto Alberto H. F. Laender Nivio Ziviani Department of Computer Science Federal University

More information

Solace JMS Broker Delivers Highest Throughput for Persistent and Non-Persistent Delivery

Solace JMS Broker Delivers Highest Throughput for Persistent and Non-Persistent Delivery Solace JMS Broker Delivers Highest Throughput for Persistent and Non-Persistent Delivery Java Message Service (JMS) is a standardized messaging interface that has become a pervasive part of the IT landscape

More information

Squid Implementing Transparent Network Caching System with Squid

Squid Implementing Transparent Network Caching System with Squid 2003 6 Squid Implementing Transparent Network Caching System with Squid lbhsieh@cc.csit.edu.tw placing tremendous demands on the Internet. A World-Wide-Web key strategy for scaling the Internet to meet

More information

Global Servers. The new masters

Global Servers. The new masters Global Servers The new masters Course so far General OS principles processes, threads, memory management OS support for networking Protocol stacks TCP/IP, Novell Netware Socket programming RPC - (NFS),

More information

MULTIMEDIA PROXY CACHING FOR VIDEO STREAMING APPLICATIONS.

MULTIMEDIA PROXY CACHING FOR VIDEO STREAMING APPLICATIONS. MULTIMEDIA PROXY CACHING FOR VIDEO STREAMING APPLICATIONS. Radhika R Dept. of Electrical Engineering, IISc, Bangalore. radhika@ee.iisc.ernet.in Lawrence Jenkins Dept. of Electrical Engineering, IISc, Bangalore.

More information

A Distributed Architecture of Edge Proxy Servers for Cooperative Transcoding

A Distributed Architecture of Edge Proxy Servers for Cooperative Transcoding A Distributed Architecture of Edge Proxy Servers for Cooperative Transcoding Valeria Cardellini University of Roma Tor Vergata cardellini@ing.uniroma2.it Michele Colajanni University of Modena colajanni@unimo.it

More information

Multiprocessors and Thread-Level Parallelism. Department of Electrical & Electronics Engineering, Amrita School of Engineering

Multiprocessors and Thread-Level Parallelism. Department of Electrical & Electronics Engineering, Amrita School of Engineering Multiprocessors and Thread-Level Parallelism Multithreading Increasing performance by ILP has the great advantage that it is reasonable transparent to the programmer, ILP can be quite limited or hard to

More information

Evaluating the Impact of Different Document Types on the Performance of Web Cache Replacement Schemes *

Evaluating the Impact of Different Document Types on the Performance of Web Cache Replacement Schemes * Evaluating the Impact of Different Document Types on the Performance of Web Cache Replacement Schemes * Christoph Lindemann and Oliver P. Waldhorst University of Dortmund Department of Computer Science

More information

Efficient Resource Management for the P2P Web Caching

Efficient Resource Management for the P2P Web Caching Efficient Resource Management for the P2P Web Caching Kyungbaek Kim and Daeyeon Park Department of Electrical Engineering & Computer Science, Division of Electrical Engineering, Korea Advanced Institute

More information

Performance and Scalability with Griddable.io

Performance and Scalability with Griddable.io Performance and Scalability with Griddable.io Executive summary Griddable.io is an industry-leading timeline-consistent synchronized data integration grid across a range of source and target data systems.

More information

An Efficient Web Cache Replacement Policy

An Efficient Web Cache Replacement Policy In the Proc. of the 9th Intl. Symp. on High Performance Computing (HiPC-3), Hyderabad, India, Dec. 23. An Efficient Web Cache Replacement Policy A. Radhika Sarma and R. Govindarajan Supercomputer Education

More information

T E C H N I C A L S A L E S S O L U T I O N S

T E C H N I C A L S A L E S S O L U T I O N S Product Management Document InterScan Web Security Virtual Appliance Customer Sizing Guide September 2010 TREND MICRO INC. 10101 N. De Anza Blvd. Cupertino, CA 95014 www.trendmicro.com Toll free: +1 800.228.5651

More information

Managing the Database

Managing the Database Slide 1 Managing the Database Objectives of the Lecture : To consider the roles of the Database Administrator. To consider the involvmentof the DBMS in the storage and handling of physical data. To appreciate

More information

The Transition to Networked Storage

The Transition to Networked Storage The Transition to Networked Storage Jim Metzler Ashton, Metzler & Associates Table of Contents 1.0 Executive Summary... 3 2.0 The Emergence of the Storage Area Network... 3 3.0 The Link Between Business

More information

WWW, REST, and Web Services

WWW, REST, and Web Services WWW, REST, and Web Services Instructor: Yongjie Zheng Aprile 18, 2017 CS 5553: Software Architecture and Design World Wide Web (WWW) What is the Web? What challenges does the Web have to address? 2 What

More information

Parallel Crawlers. Junghoo Cho University of California, Los Angeles. Hector Garcia-Molina Stanford University.

Parallel Crawlers. Junghoo Cho University of California, Los Angeles. Hector Garcia-Molina Stanford University. Parallel Crawlers Junghoo Cho University of California, Los Angeles cho@cs.ucla.edu Hector Garcia-Molina Stanford University cho@cs.stanford.edu ABSTRACT In this paper we study how we can design an effective

More information

Advanced Databases: Parallel Databases A.Poulovassilis

Advanced Databases: Parallel Databases A.Poulovassilis 1 Advanced Databases: Parallel Databases A.Poulovassilis 1 Parallel Database Architectures Parallel database systems use parallel processing techniques to achieve faster DBMS performance and handle larger

More information

Service Mesh and Microservices Networking

Service Mesh and Microservices Networking Service Mesh and Microservices Networking WHITEPAPER Service mesh and microservice networking As organizations adopt cloud infrastructure, there is a concurrent change in application architectures towards

More information

A Hybrid Load Balance Mechanism for Distributed Home Agents in Mobile IPv6

A Hybrid Load Balance Mechanism for Distributed Home Agents in Mobile IPv6 A Hybrid Load Balance Mechanism for Distributed Home Agents in Mobile IPv6 1 Hui Deng 2Xiaolong Huang 3Kai Zhang 3 Zhisheng Niu 1Masahiro Ojima 1R&D Center Hitachi (China) Ltd. Beijing 100004, China 2Dept.

More information

Improving the Database Logging Performance of the Snort Network Intrusion Detection Sensor

Improving the Database Logging Performance of the Snort Network Intrusion Detection Sensor -0- Improving the Database Logging Performance of the Snort Network Intrusion Detection Sensor Lambert Schaelicke, Matthew R. Geiger, Curt J. Freeland Department of Computer Science and Engineering University

More information

Distributed Systems. 05r. Case study: Google Cluster Architecture. Paul Krzyzanowski. Rutgers University. Fall 2016

Distributed Systems. 05r. Case study: Google Cluster Architecture. Paul Krzyzanowski. Rutgers University. Fall 2016 Distributed Systems 05r. Case study: Google Cluster Architecture Paul Krzyzanowski Rutgers University Fall 2016 1 A note about relevancy This describes the Google search cluster architecture in the mid

More information

An Cross Layer Collaborating Cache Scheme to Improve Performance of HTTP Clients in MANETs

An Cross Layer Collaborating Cache Scheme to Improve Performance of HTTP Clients in MANETs An Cross Layer Collaborating Cache Scheme to Improve Performance of HTTP Clients in MANETs Jin Liu 1, Hongmin Ren 1, Jun Wang 2, Jin Wang 2 1 College of Information Engineering, Shanghai Maritime University,

More information

An Oracle White Paper April 2010

An Oracle White Paper April 2010 An Oracle White Paper April 2010 In October 2009, NEC Corporation ( NEC ) established development guidelines and a roadmap for IT platform products to realize a next-generation IT infrastructures suited

More information

Evaluation of Strong Consistency Web Caching Techniques

Evaluation of Strong Consistency Web Caching Techniques World Wide Web: Internet and Web Information Systems, 5, 95 123, 2002 2002 Kluwer Academic Publishers. Manufactured in The Netherlands. Evaluation of Strong Consistency Web Caching Techniques L. Y. CAO

More information

HANA Performance. Efficient Speed and Scale-out for Real-time BI

HANA Performance. Efficient Speed and Scale-out for Real-time BI HANA Performance Efficient Speed and Scale-out for Real-time BI 1 HANA Performance: Efficient Speed and Scale-out for Real-time BI Introduction SAP HANA enables organizations to optimize their business

More information

From Internet Data Centers to Data Centers in the Cloud

From Internet Data Centers to Data Centers in the Cloud From Internet Data Centers to Data Centers in the Cloud This case study is a short extract from a keynote address given to the Doctoral Symposium at Middleware 2009 by Lucy Cherkasova of HP Research Labs

More information

Summary Cache: A Scalable Wide-Area Web Cache Sharing Protocol

Summary Cache: A Scalable Wide-Area Web Cache Sharing Protocol Summary Cache: A Scalable Wide-Area Web Cache Sharing Protocol Li Fan, Pei Cao and Jussara Almeida University of Wisconsin-Madison Andrei Broder Compaq/DEC System Research Center Why Web Caching One of

More information

A Statistical, Batch, Proxy-Side Web Prefetching Scheme for Efficient Internet Bandwidth Usage

A Statistical, Batch, Proxy-Side Web Prefetching Scheme for Efficient Internet Bandwidth Usage A Statistical, Batch, Proxy-Side Web Prefetching Scheme for Efficient Internet Bandwidth Usage Sook-Hyang Kim, Jae-Young Kim and James W. Hong {shk, jay, jwkhong}@postech.ac.kr Department of Computer Science

More information

Using Non-volatile Memories for Browser Performance Improvement. Seongmin KIM and Taeseok KIM *

Using Non-volatile Memories for Browser Performance Improvement. Seongmin KIM and Taeseok KIM * 2017 2nd International Conference on Computer, Network Security and Communication Engineering (CNSCE 2017) ISBN: 978-1-60595-439-4 Using Non-volatile Memories for Browser Performance Improvement Seongmin

More information

Ch. 7: Benchmarks and Performance Tests

Ch. 7: Benchmarks and Performance Tests Ch. 7: Benchmarks and Performance Tests Kenneth Mitchell School of Computing & Engineering, University of Missouri-Kansas City, Kansas City, MO 64110 Kenneth Mitchell, CS & EE dept., SCE, UMKC p. 1/3 Introduction

More information

High-Performance IP Service Node with Layer 4 to 7 Packet Processing Features

High-Performance IP Service Node with Layer 4 to 7 Packet Processing Features UDC 621.395.31:681.3 High-Performance IP Service Node with Layer 4 to 7 Packet Processing Features VTsuneo Katsuyama VAkira Hakata VMasafumi Katoh VAkira Takeyama (Manuscript received February 27, 2001)

More information

THE Lightweight Directory Access Protocol (LDAP) is

THE Lightweight Directory Access Protocol (LDAP) is IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 16, NO. 1, FEBRUARY 2008 1 Measurement and Analysis of LDAP Performance Xin Wang, Member, IEEE, Henning Schulzrinne, Fellow, IEEE, Dilip Kandlur, Fellow, IEEE,

More information

Exploiting On-Chip Data Transfers for Improving Performance of Chip-Scale Multiprocessors

Exploiting On-Chip Data Transfers for Improving Performance of Chip-Scale Multiprocessors Exploiting On-Chip Data Transfers for Improving Performance of Chip-Scale Multiprocessors G. Chen 1, M. Kandemir 1, I. Kolcu 2, and A. Choudhary 3 1 Pennsylvania State University, PA 16802, USA 2 UMIST,

More information

Data Access and Analysis with Distributed, Federated Data Servers in climateprediction.net

Data Access and Analysis with Distributed, Federated Data Servers in climateprediction.net Data Access and Analysis with Distributed, Federated Data Servers in climateprediction.net Neil Massey 1 neil.massey@comlab.ox.ac.uk Tolu Aina 2, Myles Allen 2, Carl Christensen 1, David Frame 2, Daniel

More information

Characterizing Document Types to Evaluate Web Cache Replacement Policies

Characterizing Document Types to Evaluate Web Cache Replacement Policies Characterizing Document Types to Evaluate Web Cache Replacement Policies F.J. Gonzalez-Cañete, E. Casilari, Alicia Triviño-Cabrera Dpto. Tecnología Electrónica, Universidad de Málaga, E.T.S.I. Telecomunicación,

More information

Experimental Extensions to RSVP Remote Client and One-Pass Signalling

Experimental Extensions to RSVP Remote Client and One-Pass Signalling 1 Experimental Extensions to RSVP Remote Client and One-Pass Signalling Industrial Process and System Communications, Darmstadt University of Technology Merckstr. 25 D-64283 Darmstadt Germany Martin.Karsten@KOM.tu-darmstadt.de

More information

Performance of Virtual Desktops in a VMware Infrastructure 3 Environment VMware ESX 3.5 Update 2

Performance of Virtual Desktops in a VMware Infrastructure 3 Environment VMware ESX 3.5 Update 2 Performance Study Performance of Virtual Desktops in a VMware Infrastructure 3 Environment VMware ESX 3.5 Update 2 Workload The benefits of virtualization for enterprise servers have been well documented.

More information

System Architecture PARALLEL FILE SYSTEMS

System Architecture PARALLEL FILE SYSTEMS Software and the Performance Effects of Parallel Architectures Keith F. Olsen,, Poughkeepsie, NY James T. West,, Austin, TX ABSTRACT There are a number of different parallel architectures: parallel hardware

More information

Intelligent Caching in Data Virtualization Recommended Use of Caching Controls in the Denodo Platform

Intelligent Caching in Data Virtualization Recommended Use of Caching Controls in the Denodo Platform Data Virtualization Intelligent Caching in Data Virtualization Recommended Use of Caching Controls in the Denodo Platform Introduction Caching is one of the most important capabilities of a Data Virtualization

More information

HP ProLiant BladeSystem Gen9 vs Gen8 and G7 Server Blades on Data Warehouse Workloads

HP ProLiant BladeSystem Gen9 vs Gen8 and G7 Server Blades on Data Warehouse Workloads HP ProLiant BladeSystem Gen9 vs Gen8 and G7 Server Blades on Data Warehouse Workloads Gen9 server blades give more performance per dollar for your investment. Executive Summary Information Technology (IT)

More information

Improving the Performances of Proxy Cache Replacement Policies by Considering Infrequent Objects

Improving the Performances of Proxy Cache Replacement Policies by Considering Infrequent Objects Improving the Performances of Proxy Cache Replacement Policies by Considering Infrequent Objects Hon Wai Leong Department of Computer Science National University of Singapore 3 Science Drive 2, Singapore

More information