Free upgrade of computer power with Java, web-base technology and parallel computing

Size: px
Start display at page:

Download "Free upgrade of computer power with Java, web-base technology and parallel computing"

Transcription

1 Free upgrade of computer power with Java, web-base technology and parallel computing Alfred Loo\ Y.K. Choi * and Chris Bloor* *Lingnan University, Hong Kong *City University of Hong Kong, Hong Kong ^University of Sunderland, Sunderland, United Kingdom Abstract This paper presents inexpensive ways to implement parallel computing which will make tremendous impact on many areas of computer application. Any one has knowledge of Java and more than two computers can solve their old problems with new technology by use of parallel computing. However, parallel computers are still expensive today and thus only very large companies and rich universities can afford the use of parallel computing technologies to solve their complex problems. Special training is also required for programmers who write parallel programs. With the advancement of Java and Internet/Intranet technologies, small to medium size companies can now also use the benefit of parallel computing to solve computation intensive task with their microcomputers. Most companies have a large number of microcomputers/workstations and these computers are already connected to Internet/Intranet. Additional cost for hardware and software installation is not required to prepare these computers for parallel computing. This paper presents two simple models to use ordinary microcomputers, workstations for parallel computing and a brief tutorial on the concept. Experiments are conducted to test the performance of the parallel systems. The performance of these two models will be compared. The advantages and limitations of these models are also discussed in this paper.

2 32 Applications of High-Performance Computers in Engineering VI 1 Introduction Many people view parallel computing as a rare and exotic sub-area of computing, it is interesting but of little relevance to average persons [1]. The reasons for this phenomenon are simple. Many organisations cannot afford to use parallel computers, as these machines are still too expensive. Good programmers with parallel computing training and experience are difficult to find as not all universities offer courses in parallel computing. Only rich universities can afford to buy parallel computers for these courses. However, as Internet/Intranet, Web browsers and Java technologies become ubiquitous, even small organizations will be able to use parallel computing technology to solve their data-intensive or/and computation intensive jobs. Parallel programming will become the central of the programming enterprise when people know almost every Java programmer can work on parallel computing. Parallel computing with Internet/Intranet and Java is inexpensive and easy to implement. There are no additional software and hardware installation cost. As Internet/Intranet is a popular subject, most universities offer Java courses so it will be easy to find good Java programmers. Many organisations will discover that suddenly they can now solve problems which they could not handle before. These problems need extremely powerful computers. Examples include simulation in real time, very large-scale operations research models, complex problems in artificial intelligence, etc. In this paper, we present two models for parallel computing with the Java and Intranet/Internet technologies. This model uses Java programs and a set of computers which are connected by TCP/IP protocol. Experiments are conducted to test the feasibility and efficiency of this model. The paper is organized as follows. The rest of section 1 provides the background information about the need for parallel computers and parallel algorithms. Section 2 presents a brief description on metrics in parallel systems. Section 3 presents two model of the parallel computing while the results of experiments are presented in section 4. The comparisons, advantages, disadvantages and limitations are discussed in section Driving Force for Parallel Computing Dramatic increases in computer speed have been achieved over the past forty years, but this trend will come to an end [2]. The limiting factors are the speed at which information can travel, and the distance that it has to travel within a computer. The former is limited by the speed of light. For the latter part, the technology used to reduce the size and distance between components is approaching the theoretical limit [3]. Even if we can make the distance shorter than in present technology, another problem will arise. Simultaneous electronic signal transmission between different pairs of

3 Applications of High-Performance Computers in Engineering VI 33 sequentially, speedup is limited to less than one third of what could be achievable in principle. Thus research is being focused on building efficient algorithms with few sequential operations, in order to minimise the idle time of each processor. o o Figure 1 3D plot on theoretical speedup, s, against the number of processors and fraction 2.2 Efficiency In general, only ideal systems can achieve a speedup of p for a p processors system. This implies that the fraction of operation,/, is zero. In practice, the ideal case can hardly be achieved as processors cannot devote all of their time to computing the problem as there are overheads embedded such as interprocessors communication, synchronisation, etc. Efficiency is then proposed which measures the fraction of time for which processors are usefully employed. In the ideal case, speedup is p when efficiency is one. In practice, speedup usually is less than p and efficiency is a value between zero and one. If e is the efficiency and/? is the number of processors, the formula for efficiency is as follows: efficiency speedup I p Figure 2 shows the relationship amongst p and / It is found that the efficiency drops when p and / increase. It is further found that e approaches one when / tends to zero irrespective of the number of processors in the system. This is

4 34 Applications of High-Performance Computers in Engineering VI components will start to interfere. In other words, any gains [4] in building faster electronic components will be offset by other technical difficulties. Despite these difficulties, parallel processing is still an area of growing interest to computer scientists due to enormous processing power in solving a lot of complex mathematical oriented applications such as aerodynamic simulations, image processing etc. The common factor among these applications is a "need of speed" in terms of computational requirements such as calculating a 24-hour weather forecast in less than 24 hours. There are two basic models of parallel computer systems, namely processor-to-memory model and processor-toprocessor model. A multiprocessor system consists of p processors plus interconnections for passing data and control information among the computers. Up to p different instruction streams can be active concurrently. The challenge is to put the p processors to work on different parts of a computation, so that the total computation is done at high speed by making efficient use of the processors available. The fastest supercomputer is located in the Alberta National Laboratory in United States. This computer is equivalent to 9000 Pentium processors and is able to handle 21,000 %108 instructions per second 2 Metrics in parallel systems There are many factors to assess a parallel computer's performance of which two metrics commonly used in parallel computers are Speedup and Efficiency. The details are described below. 2.1 Speedup The strongest argument against the future of parallel computing is Amdahl's Law [13] which indicates that a small number of sequential operations in a parallel algorithm can significantly limit the speedup of the whole process. The term speedup is defined as the ratio of the time required to complete the process with the fastest serial algorithm using one processor, to the time required to complete the same process with the parallel algorithm using/? processors. If / is the fraction of operations in a process which must be executed in a sequential way, then the maximum speedup which can be achieved by a computer with/? processors will be [10]: speedup <= [ 1 ] This effect is illustrated in Figure 1. Increasing the size of the number of processors, /?, and decreasing the fraction of operation, /, quickly causes the speed up to saturate. Even when only 5% of the problem is executed

5 Applications of High-Performance Computers in Engineering VI 35 expected as zero fraction time means the system is in the most efficient stage 100% devoted to computation. Figure 2 Efficiency (E) against fraction of operation, f, and the number of processors, p 3 Models for using Java and Intranet technology Many organisations have large number of personal computers and workstations. These computers are connected to their Local Area Networks under TCP/IP protocol. As Internet and Intranet are very common today, almost every personal computer has an Internet browser such as Internet Explorer and / or Netscape. Every computer with these Internet Browsers can run Java applets and they can be used for parallel computing without any additional cost. This kind of networked computers exceeds or approaches the supercomputer performance for some applications [6]. 3.1 Single Server, Multiple Clients The model is shown in Figure 3. All computers are connected by TCP/IP [7] protocol. One computer will act as a web server. A Java application program will be executed on the server. It will divide a single task (a computation extensive job) into many small sub-tasks and stored them as queues in the system. A Java applet will be stored in the server too. Other computers can join the parallel computing by clicking on the corresponding web page and download the applet. After applet reaches the destination and it will start to communicate with the application program in the server. The application program will transfer a small part of the sub-tasks to the applet and the applet can then complete the task on

6 36 Applications of High-Performance Computers in Engineering VI the client program. The computed results will then be sent to the server. Server Computer Web Server Program Application Program Client Computer 1 Client Computer i Client Computer N Figure 3 Single Server, Multiple Clients 3.2 Multiple Servers, Single Client The model is shown in Figure 4. One computer will act as a web client. A Java application program will be executed on the client. It will divide a single task (a computation extensive job) into many small sub-tasks and stored them as queues in the system. The application program invokes a servlet on the server and will transfer a small part of the sub-tasks to the servlet. The servlet can then complete the task on the server. The computed results will then be sent to the client. Client Computer #- i / j Result i* =-y Task Server Computer 1 Server Computer i Figure 4 Multiple Servers, Single Clients Server Computer N

7 Applications of High-Performance Computers in Engineering VI 37 The difference between these two models is the number of servers and clients. The former model makes use of a single server to serve the same client while the latter model uses a single client to interact with different servers. Both of them, if properly configured, can serve the purpose of parallel computation as task is decomposed into a number of sub-tasks for computation by different networked stations. 4 Experiments for the Multiple Servers Model Experiments are conducted to test the performance of the Multiple Servers Model. Experiment results are compared with our earlier studies with the Multiple Client model. 4.1 Objective of the experiments As discussed in Section 2, speedup is one of the major criteria for measuring the performance of parallel systems. The weakness of the networked computers is in the communication time between processors. As these computers are ordinary computers and they are not designed for parallel computing, the communication time between processors in our model will be longer than that of dedicated parallel computers. The objective of our experiments is to measure the speedup of our model with a good parallel algorithm. If this parallel system can achieve good speedup, then it can be used to tackle complex computing problem. 4.2 Configuration of the parallel systems A set of computers is connected with the Internet/Intranet technologies. All computers interacted with each other by stream socket connection [8] as it is more reliable than datagrams [9] techniques. A computer laboratory with over 41 microcomputers in our University was used for the experiment. All microcomputers were connected by LAN, which were used by students to practise the popular software such as Microsoft Word, Excel, Netscape, etc. in daytime. The hardware configuration for each PC is Pentium III 450 MHz PCs with 128 M RAM. Our experiments were conducted at night when no student was in the laboratory. The experiments were repeated over 30 times to ensure the data obtained are consistent. The time to complete the whole computation with different number of server computers was recorded and the results were presented below. 4.3 Result of the experiments The "speedup" is very steady even when the number of processor grows to 40 computers. The result of these two models is shown in Figures 5 and 6. It is

8 38 Applications of High-Performance Computers in Engineering VI found that the speedups are proportional to the number of processors even it is over 40. Both use an integration segment of 100 million. Figure 5 Speedup against the number of processors (1 to 10) and Figure 6 Speedup against the number of processors (10 to 40) The result of the multiple server model experiments is consistent with earlier experiment results as shown in [11] with multiple clients model. The performance of these two models are shown in Figures 7 and 8. It is revealed that increasing the number of processors will cause the increase of speedup in both cases. However, the speedup of applet on average outperforms than that of servlet and is more stable during the simulation. SerJml _ Applet i: Number of Processors Number of Processors Figure 7 Speedup against the number of processors (1 to 10) for Servlet and Applet and Figure 8 Speedup against the number of processors (10 to 40) 5 Comparison Both models achieved a linear speedup as a function of the number of processors demonstrated in section 4. It is, however, found that both share the similarities and differences, which are discussed below. 5.1 Similarities of these two modes In addition to discussion in performance, the multiple servers model shares many

9 Applications of High-Performance Computers in Engineering VI 39 advantages of multiples clients model which are discussed in our earlier publication [11]. They are described here again for readers' easy reference: Simplicity: This approach is simple and easy to implement, which means that any experienced Java programmer should be able to work on parallel computing applications with limited training. Reasonable cheap set-up cost: The technologies described above including Internet/Intranet, Java and a number of microprocessors are already available. There is no additional hardware and software cost incurred to implement such computation. Portability: The language Java used in this implementation is highly portable so it can be run on any platform. Programs developed on one platform can be run without significant modification on other platforms. Robustness: It is robust as the server shares a common bus that is proved reliably. Moreover, the client can be programmed to detect a faulty server computer and re-assign the sub task to other server to make it more reliable. 5.2 Difference of these two models Maintenance: In the multiple clients model, Java programs can be installed on the server computer. No additional technical work is needed to update the client side. It is only needed in the server side. However, in the multiple servers model, we need to update the server machine one by one. If we have a large number of servers, the maintenance job is very time consuming. Hence, the multiple clients model is better than that of multiple servers in this aspect. This drawback can be overcome by automation. It is because most web servers have the upload functions, the maintenance job can be alleviated by using this function. This can be achieved by uploading the new version of servlets to many servers from one client computer, which can be automated by use of a special script on the client computer. Operation efficiency: In the multiple clients model, we need to point to web page to start the applet on each client computer and this task is time consuming for a system with large number of computers. In the multiple servers model, invoking a servlet program on many server program can be done by executing a program on one single client computer. Hence, multiple servers model is more efficient than that of the multiple clients. 5.3 Limitation The limitation of this approach lies in the communication time which must be carefully set to enable it to be fairly decomposed into a number of sub-tasks with independable and equal computational efforts.

10 40 Applications of High-Performance Computers in Engineering VI Finally, it is expected the deployment of Java and Internet/Intranet technologies will enable small size companies to use networked computers as virtual parallel computers to solve problems. These companies will do things in a different way as they have more computing power now. Thus this proposal will have a bigger impact on the computing society. 6 References [1] I. Foster, Designing and Building Parallel Programs, Addison Wesley, 1994 [2] M. J. Quinn, Designing Efficient Algorithms for Parallel Computer, McGraw Hill, [3] A. Trew, Past, Present, Parallel; Springer Verlag, [4] S. Morse, Practical Parallel Computing, AP Professional, [5] E. Levin, Grand Challenges to computational science, Communication of ACM, vol.32, no.12, December, 1989, pp [6] C. Amza, Treadmarks; Shared Memory Computing on Networks of Workstation, IEEE Computer, July, [7] T. Parker, TCP/IP Unleashed, Sam publishing, [8] M. Campione and K. Walrath, The Java Tutorial - Object Oriented Programming for the Internet, Addision Weslet, [9] T.Courtoris, Java Networking & Communication, Prentice Hall, [10] D. Lea, Concurrent programming in Java (Addison Wesley, 1997) [11] A.Loo and C. Bloor, Parallel Computing with Java and Internet/Intranet, Microcomputer Applications, volume 18, number 3, 1999 (to be appeared).

Analytical Modeling of Parallel Programs

Analytical Modeling of Parallel Programs 2014 IJEDR Volume 2, Issue 1 ISSN: 2321-9939 Analytical Modeling of Parallel Programs Hardik K. Molia Master of Computer Engineering, Department of Computer Engineering Atmiya Institute of Technology &

More information

Evaluation of Parallel Programs by Measurement of Its Granularity

Evaluation of Parallel Programs by Measurement of Its Granularity Evaluation of Parallel Programs by Measurement of Its Granularity Jan Kwiatkowski Computer Science Department, Wroclaw University of Technology 50-370 Wroclaw, Wybrzeze Wyspianskiego 27, Poland kwiatkowski@ci-1.ci.pwr.wroc.pl

More information

Meltem Özturan misprivate.boun.edu.tr/ozturan/mis515

Meltem Özturan misprivate.boun.edu.tr/ozturan/mis515 Meltem Özturan misprivate.boun.edu.tr/ozturan/mis515 1 2 1 Selecting the Best Alternative Major Activities in the Analysis Phase Gather information Define system requirements Prototype for feasibility

More information

GRIP: A parallel graph reduction machine

GRIP: A parallel graph reduction machine GRIP: A parallel graph reduction machine Simon L. Peyton-Jones, Chris Clack and Jon Salkild University College, London Abstract GRIP - Graph Reduction In Parallel - is a Fifth Generation machine designed

More information

Intranets and Virtual Private Networks (VPNs)

Intranets and Virtual Private Networks (VPNs) Intranets and Virtual Private Networks (VPNs) Definition Private networking involves securely transmitting corporate data across multiple sites throughout an entire enterprise. Creating a truly private

More information

Unit 9 : Fundamentals of Parallel Processing

Unit 9 : Fundamentals of Parallel Processing Unit 9 : Fundamentals of Parallel Processing Lesson 1 : Types of Parallel Processing 1.1. Learning Objectives On completion of this lesson you will be able to : classify different types of parallel processing

More information

CS 590: High Performance Computing. Parallel Computer Architectures. Lab 1 Starts Today. Already posted on Canvas (under Assignment) Let s look at it

CS 590: High Performance Computing. Parallel Computer Architectures. Lab 1 Starts Today. Already posted on Canvas (under Assignment) Let s look at it Lab 1 Starts Today Already posted on Canvas (under Assignment) Let s look at it CS 590: High Performance Computing Parallel Computer Architectures Fengguang Song Department of Computer Science IUPUI 1

More information

Top500 Supercomputer list

Top500 Supercomputer list Top500 Supercomputer list Tends to represent parallel computers, so distributed systems such as SETI@Home are neglected. Does not consider storage or I/O issues Both custom designed machines and commodity

More information

Parallel Computers. c R. Leduc

Parallel Computers. c R. Leduc Parallel Computers Material based on B. Wilkinson et al., PARALLEL PROGRAMMING. Techniques and Applications Using Networked Workstations and Parallel Computers c 2002-2004 R. Leduc Why Parallel Computing?

More information

Implementation of Process Networks in Java

Implementation of Process Networks in Java Implementation of Process Networks in Java Richard S, Stevens 1, Marlene Wan, Peggy Laramie, Thomas M. Parks, Edward A. Lee DRAFT: 10 July 1997 Abstract A process network, as described by G. Kahn, is a

More information

NETWORK OF BROWSERS A MULTI-PROCESSOR COMPUTER

NETWORK OF BROWSERS A MULTI-PROCESSOR COMPUTER NETWORK OF BROWSERS A MULTI-PROCESSOR COMPUTER Luke Fletcher and Vishv Malhotra School of Computing, Private Box 1 University of Tasmania, Hobart Tas 71 Australia {luke.fletcher, vishv.malhotra} @utas.edu.au

More information

BİL 542 Parallel Computing

BİL 542 Parallel Computing BİL 542 Parallel Computing 1 Chapter 1 Parallel Programming 2 Why Use Parallel Computing? Main Reasons: Save time and/or money: In theory, throwing more resources at a task will shorten its time to completion,

More information

RAID SEMINAR REPORT /09/2004 Asha.P.M NO: 612 S7 ECE

RAID SEMINAR REPORT /09/2004 Asha.P.M NO: 612 S7 ECE RAID SEMINAR REPORT 2004 Submitted on: Submitted by: 24/09/2004 Asha.P.M NO: 612 S7 ECE CONTENTS 1. Introduction 1 2. The array and RAID controller concept 2 2.1. Mirroring 3 2.2. Parity 5 2.3. Error correcting

More information

Accelerating Implicit LS-DYNA with GPU

Accelerating Implicit LS-DYNA with GPU Accelerating Implicit LS-DYNA with GPU Yih-Yih Lin Hewlett-Packard Company Abstract A major hindrance to the widespread use of Implicit LS-DYNA is its high compute cost. This paper will show modern GPU,

More information

A Study of High Performance Computing and the Cray SV1 Supercomputer. Michael Sullivan TJHSST Class of 2004

A Study of High Performance Computing and the Cray SV1 Supercomputer. Michael Sullivan TJHSST Class of 2004 A Study of High Performance Computing and the Cray SV1 Supercomputer Michael Sullivan TJHSST Class of 2004 June 2004 0.1 Introduction A supercomputer is a device for turning compute-bound problems into

More information

Client-server Basics. 1. Motivation

Client-server Basics. 1. Motivation Page 1 of 8 Client-server Basics Oxford Brookes University 2004 Contents 1. Motivation 2. Basic Model of the Web 3. Extensions - Caching and Proxies 4. High Performance Web Servers 5. Web Services Appendices

More information

A Web Based Registration system for Higher Educational Institutions in Greece: the case of Energy Technology Department-TEI of Athens

A Web Based Registration system for Higher Educational Institutions in Greece: the case of Energy Technology Department-TEI of Athens A Web Based Registration system for Higher Educational Institutions in Greece: the case of Energy Technology Department-TEI of Athens S. ATHINEOS 1, D. KAROLIDIS 2, P. PRENTAKIS 2, M. SAMARAKOU 2 1 Department

More information

Analytical Modeling of Parallel Systems. To accompany the text ``Introduction to Parallel Computing'', Addison Wesley, 2003.

Analytical Modeling of Parallel Systems. To accompany the text ``Introduction to Parallel Computing'', Addison Wesley, 2003. Analytical Modeling of Parallel Systems To accompany the text ``Introduction to Parallel Computing'', Addison Wesley, 2003. Topic Overview Sources of Overhead in Parallel Programs Performance Metrics for

More information

High Performance Computing

High Performance Computing The Need for Parallelism High Performance Computing David McCaughan, HPC Analyst SHARCNET, University of Guelph dbm@sharcnet.ca Scientific investigation traditionally takes two forms theoretical empirical

More information

Designing Issues For Distributed Computing System: An Empirical View

Designing Issues For Distributed Computing System: An Empirical View ISSN: 2278 0211 (Online) Designing Issues For Distributed Computing System: An Empirical View Dr. S.K Gandhi, Research Guide Department of Computer Science & Engineering, AISECT University, Bhopal (M.P),

More information

Solving the N-Body Problem with the ALiCE Grid System

Solving the N-Body Problem with the ALiCE Grid System Solving the N-Body Problem with the ALiCE Grid System Dac Phuong Ho 1, Yong Meng Teo 2 and Johan Prawira Gozali 2 1 Department of Computer Network, Vietnam National University of Hanoi 144 Xuan Thuy Street,

More information

Network Definition A network can be defined as two or more computers connected together in such a way that they can share resources.

Network Definition A network can be defined as two or more computers connected together in such a way that they can share resources. Networks, telecommunications and the Internet Network Definition A network can be defined as two or more computers connected together in such a way that they can share resources. The purpose of a network

More information

Module 5 Local Area Networks

Module 5 Local Area Networks Module 5 Local Area Networks Lesson 15 LAN Topologies, Access Mechanisms And Media 5.1.1 INTRODUCTION Microelectronics brought a great revolution in computer technology. While the processing power of the

More information

Towards a Portable Cluster Computing Environment Supporting Single System Image

Towards a Portable Cluster Computing Environment Supporting Single System Image Towards a Portable Cluster Computing Environment Supporting Single System Image Tatsuya Asazu y Bernady O. Apduhan z Itsujiro Arita z Department of Artificial Intelligence Kyushu Institute of Technology

More information

Multi MicroBlaze System for Parallel Computing

Multi MicroBlaze System for Parallel Computing Multi MicroBlaze System for Parallel Computing P.HUERTA, J.CASTILLO, J.I.MÁRTINEZ, V.LÓPEZ HW/SW Codesign Group Universidad Rey Juan Carlos 28933 Móstoles, Madrid SPAIN Abstract: - Embedded systems need

More information

PRACTICAL APPROACH TO DISTRIBUTED SYSTEMS DESIGN 1

PRACTICAL APPROACH TO DISTRIBUTED SYSTEMS DESIGN 1 PRACTICAL APPROACH TO DISTRIBUTED SYSTEMS DESIGN 1 Jerzy Mieścicki, Wiktor B. Daszczuk Waldemar Grabski, Artur Krystosik Warsaw University of Technology, Institute of Computer Science {jms,wbd,wgr,akr}@ii.pw.edu.pl

More information

Multiprocessing and Scalability. A.R. Hurson Computer Science and Engineering The Pennsylvania State University

Multiprocessing and Scalability. A.R. Hurson Computer Science and Engineering The Pennsylvania State University A.R. Hurson Computer Science and Engineering The Pennsylvania State University 1 Large-scale multiprocessor systems have long held the promise of substantially higher performance than traditional uniprocessor

More information

Data and Computer Communications Chapter 1 Data Communications, Data Networks, and the Internet

Data and Computer Communications Chapter 1 Data Communications, Data Networks, and the Internet Data and Computer Communications Chapter 1 Data Communications, Data Networks, and the Internet Ninth Edition by William Stallings Data and Computer Communications, Ninth Edition by William Stallings,

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

SYSTEM UPGRADE, INC Making Good Computers Better. System Upgrade Teaches RAID

SYSTEM UPGRADE, INC Making Good Computers Better. System Upgrade Teaches RAID System Upgrade Teaches RAID In the growing computer industry we often find it difficult to keep track of the everyday changes in technology. At System Upgrade, Inc it is our goal and mission to provide

More information

Proceedings of MASPLAS'01 The Mid-Atlantic Student Workshop on Programming Languages and Systems IBM Watson Research Centre April 27, 2001

Proceedings of MASPLAS'01 The Mid-Atlantic Student Workshop on Programming Languages and Systems IBM Watson Research Centre April 27, 2001 Proceedings of MASPLAS'01 The Mid-Atlantic Student Workshop on Programming Languages and Systems IBM Watson Research Centre April 27, 2001 A PARALLEL MERGING ALGORITHM AND ITS IMPLEMENTATION WITH JAVA

More information

3/24/2014 BIT 325 PARALLEL PROCESSING ASSESSMENT. Lecture Notes:

3/24/2014 BIT 325 PARALLEL PROCESSING ASSESSMENT. Lecture Notes: BIT 325 PARALLEL PROCESSING ASSESSMENT CA 40% TESTS 30% PRESENTATIONS 10% EXAM 60% CLASS TIME TABLE SYLLUBUS & RECOMMENDED BOOKS Parallel processing Overview Clarification of parallel machines Some General

More information

Lecture 7: Parallel Processing

Lecture 7: Parallel Processing Lecture 7: Parallel Processing Introduction and motivation Architecture classification Performance evaluation Interconnection network Zebo Peng, IDA, LiTH 1 Performance Improvement Reduction of instruction

More information

Virtual private networks

Virtual private networks Technical papers Virtual private networks Virtual private networks Virtual private networks (VPNs) offer low-cost, secure, dynamic access to private networks. Such access would otherwise only be possible

More information

Introduction. Distributed Systems. Introduction. Introduction. Instructor Brian Mitchell - Brian

Introduction. Distributed Systems. Introduction. Introduction. Instructor Brian Mitchell - Brian Distributed 1 Directory 1 Cache 1 1 2 Directory 2 Cache 2 2... N Directory N Interconnection Network... Cache N N Instructor Brian Mitchell - Brian bmitchel@mcs.drexel.edu www.mcs.drexel.edu/~bmitchel

More information

Advanced Topics UNIT 2 PERFORMANCE EVALUATIONS

Advanced Topics UNIT 2 PERFORMANCE EVALUATIONS Advanced Topics UNIT 2 PERFORMANCE EVALUATIONS Structure Page Nos. 2.0 Introduction 4 2. Objectives 5 2.2 Metrics for Performance Evaluation 5 2.2. Running Time 2.2.2 Speed Up 2.2.3 Efficiency 2.3 Factors

More information

QLIKVIEW SCALABILITY BENCHMARK WHITE PAPER

QLIKVIEW SCALABILITY BENCHMARK WHITE PAPER QLIKVIEW SCALABILITY BENCHMARK WHITE PAPER Hardware Sizing Using Amazon EC2 A QlikView Scalability Center Technical White Paper June 2013 qlikview.com Table of Contents Executive Summary 3 A Challenge

More information

Future-ready IT Systems with Performance Prediction using Analytical Models

Future-ready IT Systems with Performance Prediction using Analytical Models Future-ready IT Systems with Performance Prediction using Analytical Models Madhu Tanikella Infosys Abstract Large and complex distributed software systems can impact overall software cost and risk for

More information

Metaheuristic Development Methodology. Fall 2009 Instructor: Dr. Masoud Yaghini

Metaheuristic Development Methodology. Fall 2009 Instructor: Dr. Masoud Yaghini Metaheuristic Development Methodology Fall 2009 Instructor: Dr. Masoud Yaghini Phases and Steps Phases and Steps Phase 1: Understanding Problem Step 1: State the Problem Step 2: Review of Existing Solution

More information

Insensitive Traffic Splitting in Data Networks

Insensitive Traffic Splitting in Data Networks Juha Leino and Jorma Virtamo. 2005. Insensitive traffic splitting in data networs. In: Proceedings of the 9th International Teletraffic Congress (ITC9). Beijing, China, 29 August 2 September 2005, pages

More information

Performance Evaluations for Parallel Image Filter on Multi - Core Computer using Java Threads

Performance Evaluations for Parallel Image Filter on Multi - Core Computer using Java Threads Performance Evaluations for Parallel Image Filter on Multi - Core Computer using Java s Devrim Akgün Computer Engineering of Technology Faculty, Duzce University, Duzce,Turkey ABSTRACT Developing multi

More information

COMMVAULT. Enabling high-speed WAN backups with PORTrockIT

COMMVAULT. Enabling high-speed WAN backups with PORTrockIT COMMVAULT Enabling high-speed WAN backups with PORTrockIT EXECUTIVE SUMMARY Commvault offers one of the most advanced and full-featured data protection solutions on the market, with built-in functionalities

More information

Boosting Server-to-Server Gigabit Throughput with Jumbo Frames

Boosting Server-to-Server Gigabit Throughput with Jumbo Frames Boosting Server-to-Server Gigabit Throughput with Jumbo Frames September 15, 2000 U.S.A. 2000 Hewlett-Packard Company Legal Notices The information in this document is subject to change without notice.

More information

This Lecture. BUS Computer Facilities Network Management. Switching Network. Simple Switching Network

This Lecture. BUS Computer Facilities Network Management. Switching Network. Simple Switching Network This Lecture BUS0 - Computer Facilities Network Management Switching networks Circuit switching Packet switching gram approach Virtual circuit approach Routing in switching networks Faculty of Information

More information

CHAPTER 6 MODIFIED FUZZY TECHNIQUES BASED IMAGE SEGMENTATION

CHAPTER 6 MODIFIED FUZZY TECHNIQUES BASED IMAGE SEGMENTATION CHAPTER 6 MODIFIED FUZZY TECHNIQUES BASED IMAGE SEGMENTATION 6.1 INTRODUCTION Fuzzy logic based computational techniques are becoming increasingly important in the medical image analysis arena. The significant

More information

Doctoral Written Exam in Networking, Fall 2008

Doctoral Written Exam in Networking, Fall 2008 Doctoral Written Exam in Networking, Fall 2008 December 5, 2008 Answer all parts of all questions. There are four multi-part questions, each of equal weight. Turn in your answers by Thursday, December

More information

Chapter 5: Analytical Modelling of Parallel Programs

Chapter 5: Analytical Modelling of Parallel Programs Chapter 5: Analytical Modelling of Parallel Programs Introduction to Parallel Computing, Second Edition By Ananth Grama, Anshul Gupta, George Karypis, Vipin Kumar Contents 1. Sources of Overhead in Parallel

More information

LAN Network through Power Line Communication

LAN Network through Power Line Communication International Journal of Science and Engineering Investigations vol. 4, issue 40, May 05 ISSN: 5-884 LAN Network through Power Line Communication Khalifa Dai Elnour Mohamed, Abedl Rahim ElObeid Ahmed Ph.D

More information

What is a Network? A connection of two or more computers so that they can share resources.

What is a Network? A connection of two or more computers so that they can share resources. NETWORKS What is a Network? A connection of two or more computers so that they can share resources. Network Benefits Remote access Sharing files & resources Communication Cost Maintenance Communication

More information

Parallel Programming Concepts. Parallel Algorithms. Peter Tröger

Parallel Programming Concepts. Parallel Algorithms. Peter Tröger Parallel Programming Concepts Parallel Algorithms Peter Tröger Sources: Ian Foster. Designing and Building Parallel Programs. Addison-Wesley. 1995. Mattson, Timothy G.; S, Beverly A.; ers,; Massingill,

More information

Computer Architecture

Computer Architecture Computer Architecture Chapter 7 Parallel Processing 1 Parallelism Instruction-level parallelism (Ch.6) pipeline superscalar latency issues hazards Processor-level parallelism (Ch.7) array/vector of processors

More information

Data and Computer Communications

Data and Computer Communications Data and Computer Communications CHAPTER 1 Tenth Edition by William Stallings Data Communications, Data Networks, and the Internet Data and Computer Communications, Tenth Edition by William Stallings,

More information

Performance Analysis and Optimal Utilization of Inter-Process Communications on Commodity Clusters

Performance Analysis and Optimal Utilization of Inter-Process Communications on Commodity Clusters Performance Analysis and Optimal Utilization of Inter-Process Communications on Commodity Yili TSENG Department of Computer Systems Technology North Carolina A & T State University Greensboro, NC 27411,

More information

Parallel Combinatorial Search on Computer Cluster: Sam Loyd s Puzzle

Parallel Combinatorial Search on Computer Cluster: Sam Loyd s Puzzle Parallel Combinatorial Search on Computer Cluster: Sam Loyd s Puzzle Plamenka Borovska Abstract: The paper investigates the efficiency of parallel branch-and-bound search on multicomputer cluster for the

More information

CSC630/CSC730 Parallel & Distributed Computing

CSC630/CSC730 Parallel & Distributed Computing CSC630/CSC730 Parallel & Distributed Computing Analytical Modeling of Parallel Programs Chapter 5 1 Contents Sources of Parallel Overhead Performance Metrics Granularity and Data Mapping Scalability 2

More information

Chapter 13 Strong Scaling

Chapter 13 Strong Scaling Chapter 13 Strong Scaling Part I. Preliminaries Part II. Tightly Coupled Multicore Chapter 6. Parallel Loops Chapter 7. Parallel Loop Schedules Chapter 8. Parallel Reduction Chapter 9. Reduction Variables

More information

Introduction to Parallel Programming

Introduction to Parallel Programming Introduction to Parallel Programming January 14, 2015 www.cac.cornell.edu What is Parallel Programming? Theoretically a very simple concept Use more than one processor to complete a task Operationally

More information

Tools for Remote Web Usability Evaluation

Tools for Remote Web Usability Evaluation Tools for Remote Web Usability Evaluation Fabio Paternò ISTI-CNR Via G.Moruzzi, 1 56100 Pisa - Italy f.paterno@cnuce.cnr.it Abstract The dissemination of Web applications is enormous and still growing.

More information

COMPUTER ORGANISATION CHAPTER 1 BASIC STRUCTURE OF COMPUTERS

COMPUTER ORGANISATION CHAPTER 1 BASIC STRUCTURE OF COMPUTERS Computer types: - COMPUTER ORGANISATION CHAPTER 1 BASIC STRUCTURE OF COMPUTERS A computer can be defined as a fast electronic calculating machine that accepts the (data) digitized input information process

More information

Use of Tree-based Algorithms for Internal Sorting

Use of Tree-based Algorithms for Internal Sorting Use of Tree-based s for Internal Sorting P. Y. Padhye (puru@deakin.edu.au) School of Computing and Mathematics Deakin University Geelong, Victoria 3217 Abstract Most of the large number of internal sorting

More information

1.0 INTRODUCTION. Introduction to data communication and networking 1

1.0 INTRODUCTION. Introduction to data communication and networking 1 Introduction to data communication and networking 1 1.0 INTRODUCTION Data communications and networking may be the fastest growing technologies in our culture today. One of the ramifications of that growth

More information

Technical Documentation Version 7.4. Performance

Technical Documentation Version 7.4. Performance Technical Documentation Version 7.4 These documents are copyrighted by the Regents of the University of Colorado. No part of this document may be reproduced, stored in a retrieval system, or transmitted

More information

City University of Hong Kong Course Syllabus. offered by Department of Computer Science with effect from Semester A 2017/18

City University of Hong Kong Course Syllabus. offered by Department of Computer Science with effect from Semester A 2017/18 City University of Hong Kong offered by Department of Computer Science with effect from Semester A 2017/18 Part I Course Overview Course Title: Computer Networks Course Code: CS3201 Course Duration: 1

More information

1 Introduction to Parallel Computing

1 Introduction to Parallel Computing 1 Introduction to Parallel Computing 1.1 Goals of Parallel vs Distributed Computing Distributed computing, commonly represented by distributed services such as the world wide web, is one form of computational

More information

The Need for Speed: Understanding design factors that make multicore parallel simulations efficient

The Need for Speed: Understanding design factors that make multicore parallel simulations efficient The Need for Speed: Understanding design factors that make multicore parallel simulations efficient Shobana Sudhakar Design & Verification Technology Mentor Graphics Wilsonville, OR shobana_sudhakar@mentor.com

More information

Software Development Chapter 1

Software Development Chapter 1 Software Development Chapter 1 1. Introduction Software Applications are increasingly used to tackle problems that concern everyday life : Automatic Bank tellers Airline reservation systems Air traffic

More information

Parallelizing TCP/IP Offline Log Analysis and Processing Exploiting Multiprocessor Functionality

Parallelizing TCP/IP Offline Log Analysis and Processing Exploiting Multiprocessor Functionality Parallelizing TCP/IP Offline Log Analysis and Processing Exploiting Multiprocessor Functionality Chirag Kharwar Department Of Computer Science & Engineering Nirma university Abstract In the era of internet

More information

COST ESTIMATION FOR DISTRIBUTED SYSTEMS USING USE CASE DIAGRAM

COST ESTIMATION FOR DISTRIBUTED SYSTEMS USING USE CASE DIAGRAM S. V. Pingale et al. : Cost Estimation for Distributed Systems using Use Case Diagram Journal of Advances in Engineering Science 41 Section C (3), July - December 2010, PP 41-48 COST ESTIMATION FOR DISTRIBUTED

More information

Lecture 7: Parallel Processing

Lecture 7: Parallel Processing Lecture 7: Parallel Processing Introduction and motivation Architecture classification Performance evaluation Interconnection network Zebo Peng, IDA, LiTH 1 Performance Improvement Reduction of instruction

More information

Implementation of Classification Rules using Oracle PL/SQL

Implementation of Classification Rules using Oracle PL/SQL 1 Implementation of Classification Rules using Oracle PL/SQL David Taniar 1 Gillian D cruz 1 J. Wenny Rahayu 2 1 School of Business Systems, Monash University, Australia Email: David.Taniar@infotech.monash.edu.au

More information

Parallel Programming Programowanie równoległe

Parallel Programming Programowanie równoległe Parallel Programming Programowanie równoległe Lecture 1: Introduction. Basic notions of parallel processing Paweł Rzążewski Grading laboratories (4 tasks, each for 3-4 weeks) total 50 points, final test

More information

Copyright 2009 by Scholastic Inc. All rights reserved. Published by Scholastic Inc. PDF0090 (PDF)

Copyright 2009 by Scholastic Inc. All rights reserved. Published by Scholastic Inc. PDF0090 (PDF) Enterprise Edition Version 1.9 System Requirements and Technology Overview The Scholastic Achievement Manager (SAM) is the learning management system and technology platform for all Scholastic Enterprise

More information

Overview of CentreWare Page 1 of 6. CentreWare Overview

Overview of CentreWare Page 1 of 6. CentreWare Overview Overview of CentreWare Page 1 of 6 CentreWare Overview A Software Portfolio for Managing Documents, Network Printers, and Peripheral Devices Across Diverse Client/Server Network Environments April 1997

More information

SUMMERY, CONCLUSIONS AND FUTURE WORK

SUMMERY, CONCLUSIONS AND FUTURE WORK Chapter - 6 SUMMERY, CONCLUSIONS AND FUTURE WORK The entire Research Work on On-Demand Routing in Multi-Hop Wireless Mobile Ad hoc Networks has been presented in simplified and easy-to-read form in six

More information

MATRIX-VECTOR MULTIPLICATION ALGORITHM BEHAVIOR IN THE CLOUD

MATRIX-VECTOR MULTIPLICATION ALGORITHM BEHAVIOR IN THE CLOUD ICIT 2013 The 6 th International Conference on Information Technology MATRIX-VECTOR MULTIPLICATIO ALGORITHM BEHAVIOR I THE CLOUD Sasko Ristov, Marjan Gusev and Goran Velkoski Ss. Cyril and Methodius University,

More information

6.2 DATA DISTRIBUTION AND EXPERIMENT DETAILS

6.2 DATA DISTRIBUTION AND EXPERIMENT DETAILS Chapter 6 Indexing Results 6. INTRODUCTION The generation of inverted indexes for text databases is a computationally intensive process that requires the exclusive use of processing resources for long

More information

Utilizing a Common Language as a Generative Software Reuse Tool

Utilizing a Common Language as a Generative Software Reuse Tool Utilizing a Common Language as a Generative Software Reuse Tool Chris Henry and Stanislaw Jarzabek Department of Computer Science School of Computing, National University of Singapore 3 Science Drive,

More information

Using implicit fitness functions for genetic algorithm-based agent scheduling

Using implicit fitness functions for genetic algorithm-based agent scheduling Using implicit fitness functions for genetic algorithm-based agent scheduling Sankaran Prashanth, Daniel Andresen Department of Computing and Information Sciences Kansas State University Manhattan, KS

More information

Development of a Web based Education Module for Verilog HDL

Development of a Web based Education Module for Verilog HDL Proceedings of the 2010 International Conference on Industrial Engineering and Operations Management Dhaka, Bangladesh, January 9 10, 2010 Development of a Web based Education Module for Verilog HDL Rifat

More information

SAS Meets Big Iron: High Performance Computing in SAS Analytic Procedures

SAS Meets Big Iron: High Performance Computing in SAS Analytic Procedures SAS Meets Big Iron: High Performance Computing in SAS Analytic Procedures Robert A. Cohen SAS Institute Inc. Cary, North Carolina, USA Abstract Version 9targets the heavy-duty analytic procedures in SAS

More information

Chapter 7. Multicores, Multiprocessors, and Clusters. Goal: connecting multiple computers to get higher performance

Chapter 7. Multicores, Multiprocessors, and Clusters. Goal: connecting multiple computers to get higher performance Chapter 7 Multicores, Multiprocessors, and Clusters Introduction Goal: connecting multiple computers to get higher performance Multiprocessors Scalability, availability, power efficiency Job-level (process-level)

More information

The Art of Parallel Processing

The Art of Parallel Processing The Art of Parallel Processing Ahmad Siavashi April 2017 The Software Crisis As long as there were no machines, programming was no problem at all; when we had a few weak computers, programming became a

More information

Computer Overview. A computer item you can physically see or touch. A computer program that tells computer hardware how to operate.

Computer Overview. A computer item you can physically see or touch. A computer program that tells computer hardware how to operate. Hardware Computer Overview A computer item you can physically see or touch. Software A computer program that tells computer hardware how to operate. Information Technology (IT) The broad subject related

More information

Exploring different level of parallelism Instruction-level parallelism (ILP): how many of the operations/instructions in a computer program can be performed simultaneously 1. e = a + b 2. f = c + d 3.

More information

Chapter One. Computers in Context

Chapter One. Computers in Context Chapter One Computers in Context After reading this chapter, you should be able to: Characterize what a computer is and what it does Describe several ways computers play a critical role in modern life

More information

Interconnect Technology and Computational Speed

Interconnect Technology and Computational Speed Interconnect Technology and Computational Speed From Chapter 1 of B. Wilkinson et al., PARAL- LEL PROGRAMMING. Techniques and Applications Using Networked Workstations and Parallel Computers, augmented

More information

Continuous Real Time Data Transfer with UDP/IP

Continuous Real Time Data Transfer with UDP/IP Continuous Real Time Data Transfer with UDP/IP 1 Emil Farkas and 2 Iuliu Szekely 1 Wiener Strasse 27 Leopoldsdorf I. M., A-2285, Austria, farkas_emil@yahoo.com 2 Transilvania University of Brasov, Eroilor

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

Pervasive.SQL - Configuration & Performance Settings

Pervasive.SQL - Configuration & Performance Settings Pervasive.SQL - Configuration & Performance Settings Neil Hughes Support4Omega Ltd. 19th June 2005 1 Introduction Although one of Pervasive.SQL s advantages is its ability to be installed and run straight

More information

ENSC 427: COMMUNICATION NETWORKS (Spring 2011) Final Report

ENSC 427: COMMUNICATION NETWORKS (Spring 2011) Final Report ENSC 427: COMMUNICATION NETWORKS (Spring 2011) Final Report Video Streaming over the 802.11g WLAN Technologies http://www.sfu.ca/~zxa7/ Zhenpeng Xue 301062408 zxa7@sfu.ca Page 2 of 16 Table of Contents

More information

Considering the 2.5-inch SSD-based RAID Solution:

Considering the 2.5-inch SSD-based RAID Solution: Considering the 2.5-inch SSD-based RAID Solution: Using Infortrend EonStor B12 Series with Intel SSD in a Microsoft SQL Server Environment Application Note Abstract This application note discusses the

More information

Advanced Topics in Computer Architecture

Advanced Topics in Computer Architecture Advanced Topics in Computer Architecture Lecture 7 Data Level Parallelism: Vector Processors Marenglen Biba Department of Computer Science University of New York Tirana Cray I m certainly not inventing

More information

MPI Optimisation. Advanced Parallel Programming. David Henty, Iain Bethune, Dan Holmes EPCC, University of Edinburgh

MPI Optimisation. Advanced Parallel Programming. David Henty, Iain Bethune, Dan Holmes EPCC, University of Edinburgh MPI Optimisation Advanced Parallel Programming David Henty, Iain Bethune, Dan Holmes EPCC, University of Edinburgh Overview Can divide overheads up into four main categories: Lack of parallelism Load imbalance

More information

MEASURING COMPUTER TIME. A computer faster than another? Necessity of evaluation computer performance

MEASURING COMPUTER TIME. A computer faster than another? Necessity of evaluation computer performance Necessity of evaluation computer performance MEASURING COMPUTER PERFORMANCE For comparing different computer performances User: Interested in reducing the execution time (response time) of a task. Computer

More information

E-Commerce. Infrastructure I: Computer Networks

E-Commerce. Infrastructure I: Computer Networks E-Commerce Infrastructure I: Computer Networks Almost all computers today are networked or part of a distributed system. I will provide an overview of networking and a basic description of network technology.

More information

2. Give advantages and disadvantages of computer network. (Any 2 advantages- 2 Marks, Any 2 disadvantages- 2 Marks) Answer: Advantages:

2. Give advantages and disadvantages of computer network. (Any 2 advantages- 2 Marks, Any 2 disadvantages- 2 Marks) Answer: Advantages: Summer-15 1. Compare human network and computer network. (Each point- 1 Mark, any two points) Human Network 1. Human Network is a network of human beings. 2.It consist of individuals,organizations, schools,

More information

A Distance Learning Tool for Teaching Parallel Computing 1

A Distance Learning Tool for Teaching Parallel Computing 1 A Distance Learning Tool for Teaching Parallel Computing 1 RAFAEL TIMÓTEO DE SOUSA JR., ALEXANDRE DE ARAÚJO MARTINS, GUSTAVO LUCHINE ISHIHARA, RICARDO STACIARINI PUTTINI, ROBSON DE OLIVEIRA ALBUQUERQUE

More information

Introduction to Networking

Introduction to Networking Introduction to Networking The fundamental purpose of data communications is to exchange information between user's computers, terminals and applications programs. Simplified Communications System Block

More information

Middle East Technical University. Department of Computer Engineering

Middle East Technical University. Department of Computer Engineering Middle East Technical University Department of Computer Engineering TurkHITs Software Requirements Specifications v1.1 Group fourbytes Safa Öz - 1679463 Mert Bahadır - 1745785 Özge Çevik - 1679414 Sema

More information

Applications and Performance Analysis of Bridging with L3 Forwarding on Wireless LANs

Applications and Performance Analysis of Bridging with L3 Forwarding on Wireless LANs Applications and Performance Analysis of Bridging with L3 Forwarding on Wireless LANs Chibiao Liu and James Yu DePaul University School of CTI Chicago, IL {cliu1, jyu}@cs.depaul.edu Abstract This paper

More information