Task Distribution in a Workstation Cluster with a Concurrent Network

Size: px
Start display at page:

Download "Task Distribution in a Workstation Cluster with a Concurrent Network"

Transcription

1 Task Distribution in a Workstation Cluster with a Concurrent Network Frank Burchert, Michael Koch, Gunther Hipper, Djamshid Tavangarian Universität Rostock, Fachbereich Informatik, Institut für Technische Informatik Albert-Einstein-Straße 21, D Rostock (Germany) Tel.: ++49 (381) , Fax: ++49 (381) Burchert@Informatik.Uni-Rostock.de Abstract This paper concentrates on task allocation and load balancing within a Concurrent Network Architecture Cluster using simple strategies for distributing tasks. A sender initiated approach, a receiver initiated approach with reservation strategy, a deterministic Greedy algorithm similar to the sender initiated method and a gradient algorithm adapted to the special communication possibilities of CNA are investigated and compared. First results gained by simulation show that the gradient model load balancing method yields the best increase in performance. In addition, simple methods used to get parameters required by these load balancing algorithms in a UNIX environment are introduced. Key words distributed systems, load balancing, load sharing, workstation cluster 1. Introduction Over the past few years, claims respecting the performance of computer systems have grown steadily. In order to satisfy these claims, new techniques and architectures are necessary for both soft- and hardware. Parallel computing has proved to be a suitable tool, but requires specialized parallel hardware structures. The use of workstation clusters (C) as distributed memory parallel computers, which has gained increasing acceptance, has proved to be a suitable and cost-effective approach. The main bottleneck of a distributed memory computer architecture is the performance of the communication system connecting the processing elements. Conventional clusters consist of a single, shared communication channel which becomes disadvantageous if the amount of exchanged data exceeds the usual share for which the network was primary dimensioned (e.g. file service). Future LAN technologies will be able to reduce this bottleneck but not to eliminate it. A modified communication architecture called Concurrent Network Architecture (CNA) extends the concurrency of the processing elements to their communication network [1]. The following chapter examines the basic idea of CNA. Another approach used to increase system performance can be found in the application of load balancing and load sharing strategies to C. Over the last decade, many load balancing algorithms have been developed in order to equalize system load over all workstations () within a network [2]. Load balancing prevents some workstations from staying idle while others within the cluster are under heavy load. Therefore, the length of time during which the workstations are idle is reduced and the system performance is increased. This paper concentrates on task allocation and load balancing within a CNA cluster using simple strategies for distributing serial tasks. With respect to the special CNA communication features, the following strategies are compared: a sender initiated approach [3], a receiver initiated approach with reservation strategy [3], a deterministic Greedy algorithm similar to the sender initiated method [4] and a gradient algorithm adapted to the special communication possibilities of CNA [5]. First Sino-German Workshop on Advanced Parallel Processing Technologies, APPT 95, Beijing, China, September / 7

2 The first chapter introduces the concept of the Concurrent Network Architecture. In the second chapter, different load balancing algorithms and their parameters are demonstrated. The third chapter concentrates on methods used to determine these parameters in a real, heterogeneous UNIX environment. In the final chapter, simulation results concerning the improvement of job throughput gained by load balancing and sharing within a CNA cluster are presented. The paper ends with conclusions and a description of future work. 2. The Concurrent Network Architecture Concurrent computing in workstation clusters as an alternative to MIMD- or SIMD-computer architectures has gained increasing acceptance within the last few years. The main reason for this is the superior cost / performance ratio of a C compared to conventional parallel architectures. In many places, workstations are already connected by a LAN which is sufficient for usual file server activities. Such a workstation cluster can be regarded as a distributed memory parallel computer. In principal, clusters with a single communication channel (denoted as conventional clusters) can be used to execute concurrent applications. Research has proved that the network often becomes a bottleneck which subsequently reduces system performance. In order to avoid this, a new communication architecture called CNA (Concurrent Network Architecture) is proposed in [1] through which the concurreny of the processing elements to the network is extended. An economic solution which is able to enhance the communication performance whilst maintaining the cost-effectiveness of C can be found by modifying existing networks. Within a CNA Cluster, workstations () are structured in a regular n-dimensional mesh. Every is connected to n different and independent communication network channels (e.g. Ethernet, FDDI, ATM). Each has two different IP-addresses and each CPU of the has to perform routing tasks of the IP-protocols between its communication channels [6]. a) b) ,3 2,3 3, ,2 2,2 3, ,1 2,1 3,1 Conventional cluster with a single communication channel Example of a two-dimensional 3x3 CNA cluster Figure 1: Conventional cluster architecture vs. Concurrent Network Architecture In order to build a CNA cluster, both additional communication hardware for the implementation of the concurrency of the communication system (see fig. 1) and software components are needed. A part of these components are the TCP/IP and UDP/IP protocol stacks as an element of the UNIX operating system. Another part of this software is an optimized programming environment for the CNA cluster, which is based on a message-passing model. These programs and libraries enable the programmer of a parallel application to use the special feature of the CNA, this being independent communication channels which realize a higher system bandwidth. A realization of such a communication software is described in [6]. First Sino-German Workshop on Advanced Parallel Processing Technologies, APPT 95, Beijing, China, September / 7

3 Fig. 2 shows examples of possible routing ways within a CNA cluster. There are different possibilities for sending messages from one to another which depends on the load of each communication channel. It is also possible to divide the message into halves and transmit each half along an own channel. By this, the bandwidth between two integrated in the CNA cluster is extended. 1,3 2,3 3,3 Y 3 Patterns of independent communications between : Y 2 1,2 2,2 3,2 a) (3,1) and (3,2) b) (1,1) and (2,1) Y 1 1,1 2,1 3,1 c) (1,3) and (2,2) X 1 X 2 X 3 Figure 2: Example of independent communications in a two-dimensional CNA cluster 3. Load balancing strategies Load balancing strategies can be divided into two main categories: static policies which allocate tasks only before they have started and dynamic methods which also reallocate tasks at run time. The benefit of dynamic load balancing is evaluated controversially in literature. Some authors prove the theory that dynamic load balancing offers additional performance [9][10][11][12], whereas others prove that the overhead reaches or exceeds the benefits [13][14][15]. The emphasis of this paper will be on static strategies, due to their simple implementability. If arbitrary tasks are scheduled in a CNA cluster, one has to distinguish between scheduling parallel and sequential tasks. Usually, parallel tasks produce much more traffic on the communication channel than sequential tasks. Hence, to allocate parallel tasks, algorithms are needed which especially take communication costs between the single modules into account. Several load balancing algorithms are based on graph theory [7][8]. The basic ideas are presented in fig. 3: Tasks t i (and in Stone s approach also the processing elements P k ) are represented by nodes. The edges connecting the nodes are weighted with communication costs c ij (edges between two task nodes) or with specially-defined execution costs w ij (edges between processes and processors) denoting the costs if task t i is not executed on P j. Through the use of the Ford-Fulkerson-Algorithm, task allocations which cause minimum costs can be obtained. n-way cut Figure 3: Task allocation by use of min cut / max flow algorithm [7] First Sino-German Workshop on Advanced Parallel Processing Technologies, APPT 95, Beijing, China, September / 7

4 This algorithm yields optimal task allocations only for two-processor-systems. If there are more than two processing elements, the policy becomes much more complex since the Ford-Fulkerson-Algorithm has to be performed numerous times. In addition, optimal allocations are not guaranteed. Therefore, Lo propagates an extension with a heuristic approach [8] which leads to suboptimal distributions. This is a suitable basis for efficient scheduling of parallel tasks within CNA clusters. If sequential tasks have to be distributed within a cluster, uncomplicated strategies are sufficient to distinctly improve performance. Table 1 shows some simple algorithms proposed in [3][4][5], which were adapted to CNA and its special communication capabilities. Therefore, the definition of immediate neighbours within the gradient model was extended to which are directly connected by a common communication channel. The applied load index is the amount of processes running on a. Although this is a simple load index, Eager et al. proved in [16] that a distinct increase of the performance is possible. Extensions to this load index contain the used main memory and swap space and cpu use. name sender-initiated receiver-initiated reservation method greedy algorithm gradient method short description Table 1: Simple load balancing strategies with threshold policies 4. Detecting load parameters under UNIX heavy-loaded nodes try to find lower-loaded nodes by incident and then transmit the newly-arrived task [3] low-loaded nodes try to find heavy loaded nodes by incident and then transmit the newly-arrived task [3] nodes at which tasks have just terminated reserve for the next arriving task on a heavy-loaded node to transmit it on arrival [3] similar to the sender-initiated algorithm, but with a deterministic choice of target nodes [4] a newly-arrived task is sent to the lowest loaded node in the neighbourhood (up to n hops) [5] Load balancing in homogeneous environments is a well investigated topic, but difficulties appear if strategies have to take heterogeneity into account. The complexity of these algorithms and the fact that adapted software is needed for detecting the required parameters impede general implementations of load balancing strategies for heterogeneous clusters. UNIX systems offer a number of user commands, maintenance commands and system calls for information about system and process state. Some of them are listed in table 2. The fact that there is no common standard for UNIX process- or system information makes it difficult to implement universal programs for monitoring the most interesting system parameters. Moreover, professional monitoring programs do not yield a uniform interface where required parameters can be received. In general, it would be useful to take only commands and system calls which exist on any machine and guarantee the same output. However, calling these commands frequently initiates fork-execcombinations which extremely increase the load at the moment of measurement and falsify system parameters like cpu use and memory accesses [2]. Hence, other methods are needed to get information about the system state in a single process. First Sino-German Workshop on Advanced Parallel Processing Technologies, APPT 95, Beijing, China, September / 7

5 name user command system call maintenance command LINUX SunOS System V description ps X X X X report process status pstat X X print system facts iostat X X X report i/o statistics vmstat X (X) X X report virtual memory statistics w X (X) X X who is logged in and what they are doing uname X X X X X system identification sysinfo X X X *) system information netstat X X X show network status top X X system and process information time X X X X time a command times X X X X get topical process times /proc X X process file system acct... X (X) X X process accounting sa X X X system accounting *) different meaning under System V Table 2: General sources of system and process information under several UNIX versions Newer UNIX versions offer information on the most important system facts within the process file system /proc maintained by the kernel in main memory. Unfortunately, even here no standard is in sight. The volume of offered information differs as well as forms of representation. LINUX, for instance, offers system and process information in the form of ASCII files, whereas Solaris yields pointers to special memory areas where the information is stored. Scheduler uniform load information basis LINUX SunOS System V Solaris OSF/1... Figure 4: The necessity of an uniform load information basis First Sino-German Workshop on Advanced Parallel Processing Technologies, APPT 95, Beijing, China, September / 7

6 A uniform basis for load and system facts which provides the scheduler with required information (see fig. 4) is useful for a general implementation of load balancing policies in a heterogeneous environment. First realizations of such a uniform load information basis exist for SunOS4 and LINUX. Under SunOS4, direct access to the process table is necessary, whereas under LINUX the process file system is used in combination with system calls. Further development will contain other operating systems. 5. Simulation model and first results An event-driven simulation was used to compare the behaviour of the algorithms described in table 2 within an 3x3-CNA cluster using ethernet communication channels. Within this model, workstations are represented by dynamic process lists, and the processing of the inserted events is individually controlled for each depending on the length of its process list (load depending characteristic). The predefined process times (time the processes have to stay in the system) are increased depending on the final location of the process within the cluster if no local processing is instructed by the scheduler. Communication costs and the time interval determining the load measurement period are varied. In order to investigate the behaviour of these algorithms in a heterogeneous environment, performance factors were used to differentiate between more and less powerful within the cluster. Fig. 5a and 5b show the behaviour of the tested policies at different transfer costs and the effect of the update period for load information. The figures show the mean response time of the simulated processes standardized on the mean response time when no load balancing is performed. Transfer costs refer to direct transfers without any gateway. In simulation, these costs increase to twice the value if the transfer is routed along a gateway (recent research has shown that the real factor in the CNA cluster is ca. 1.5 [6], improving the results). mean response time (standardized) None = Sender Greedy 0.7 Gradient Reserv transfer costs (in % of processing costs) a) effect of different transfer costs mean response time (standardized) None = 1 Sender Greedy Gradient Reserv update period (in abstract time units) b) effect of the update period for load information Figure 5: The reduction of the response time depending on transfer costs and update period for load information The gradient strategy yields the shortest mean response time if the update period for load information is short enough compared to the global task arrival rate which was defined at 10 time units. For update periods much longer than the global arrival rate, the information basis of the scheduler does not correspond to the real load in the system, so that the mean response time even exceeds the value received without scheduling. Although the afore-mentioned strategies originally were created for homogeneous environments, further investigation has shown that they also yield proper performance improvements within heterogeneous environments. This can be explained due to the special load index which enables more powerful machines to be indirectly recognized since the higher processing performance leads to more task terminations within an update period than on other machines. First Sino-German Workshop on Advanced Parallel Processing Technologies, APPT 95, Beijing, China, September / 7

7 6. Conclusions This paper presented the concept of a CNA cluster and introduced static load balancing strategies suitable for distributing arbitrary tasks within such an environment. Different approaches were proposed to distribute parallel and sequential tasks. With respect to sequential tasks, first results show that the gradient model load balancing method is superior to sender-initiated and receiver-initiated strategies if it is adapted to the special communication possibilities. Future work will concentrate on simulating the proposed load balancing algorithm for distributing parallel tasks and extending the uniform load information basis to other UNIX versions like System V. The aim is the development of a comprehensive scheduling system which enables the use of CNA clusters as general servers for arbitrary tasks. References [1] Klein; Tavangarian; Hipper; Koch: A New Concurrent Network Architecture (CNA) for an Efficient Parallel Computing in Workstation Clusters, Workshop Parallele Datenverarbeitung im Verbund von Hochleistungs-Workstations, Fachberichte Informatik, 09/1994, Universität Koblenz- Landau [2] Burchert: Verteilung von Applikationen in einem konzentrierten Workstation-Cluster, Diplomarbeit am Lehrgebiet Technische Informatik II, FernUniversität Hagen, 1995 [3] Eager; Lazowska; Zahorjan: A Comparison of Receiver-Initiated and Sender-Initiated Adaptive Load Sharing, Performance Evaluation, Vol. 6, No. 1 03/1986, p [4] Chowdhury: The Greedy Load Sharing Algorithm, Journal of Parallel and Distributed Computing (Academic Press), No. 9, 1990, p [5] Lin; Keller: The Gradient Model Load Balancing Method, IEEE Transactions on Software Engineering, Vol. SE-13, No. 1, 01/1987, p [6] Hipper; Koch; Tavangarian: Eine parallele Kommunikationsarchitektur für Workstation-Cluster, 3. GI/ITG Fachtagung Arbeitsplatzrechensysteme (APS), Hannover 1995, ISBN , p [7] Stone: Multiprocessor Scheduling with the Aid of Network Flow Algorithms, IEEE Transactions on Software Engineering, Vol. SE-3, No. 1, 01/1977, P [8] Lo, V. M.: Heuristic Algorithms for Task Assignment in Distributed Systems, IEEE Transactions on Computers, Vol. 37, No. 11, 11/1988, p [9] Baraq, Annon et al.: The MOSIX distributed operating system - Load Balancing for UNIX, Springer 1993, ISBN [10] Douglis, Fred; Ousterhout, John: Transparent Process Migration: Design Alternatives and the Sprite Implementation, Software - Practice and Experience, Vol. 21, No. 8, 08/1991, p [11] Krueger; Livny: A Comparison of Preemptive and Non-Preemptive Load Distributing, Proceedings of the 8th International Conference on Distributed Computing Systems (IEEE), 06/1988, p [12] Beerbohm; Bresgen; Hofestädt; Huang: Dynamic Load Balancing on Workstation Clusters, Workshop Parallele Datenverarbeitung im Verbund von Hochleistungs-Workstations, Fachberichte Informatik, 09/1994, Universität Koblenz-Landau [13] Eager; Lazowska; Zahorjan: The Limited Performance Benefits of Migrating Active Processes for Load Sharing, ACM Performance Evaluation Review, Vol. 16, No. 1, 05/1988, p [14] Joosen; Verbaeten: On the Use of Process Migration in Distributed Systems, Microprocessing and Microprogramming (North-Holland), 1986, p [15] Leland; Ott; Teunis: Load Balancing Heuristics and Process Behavior, ACM Performance Evaluation Review, Vol. C-33, No. 1, 05/1986, p [16] Eager; Lazowska; Zahorjan: Adaptive Load Sharing in Homogeneous Distributed Systems, IEEE Transactions on Software Engineering, Vol. SE-12, No. 5, 05/1986, p First Sino-German Workshop on Advanced Parallel Processing Technologies, APPT 95, Beijing, China, September / 7

Chapter 5: Distributed Process Scheduling. Ju Wang, 2003 Fall Virginia Commonwealth University

Chapter 5: Distributed Process Scheduling. Ju Wang, 2003 Fall Virginia Commonwealth University Chapter 5: Distributed Process Scheduling CMSC 602 Advanced Operating Systems Static Process Scheduling Dynamic Load Sharing and Balancing Real-Time Scheduling Section 5.2, 5.3, and 5.5 Additional reading:

More information

Dynamic Balancing Complex Workload in Workstation Networks - Challenge, Concepts and Experience

Dynamic Balancing Complex Workload in Workstation Networks - Challenge, Concepts and Experience Dynamic Balancing Complex Workload in Workstation Networks - Challenge, Concepts and Experience Abstract Wolfgang Becker Institute of Parallel and Distributed High-Performance Systems (IPVR) University

More information

A Distributed System with a Centralized Organization

A Distributed System with a Centralized Organization A Distributed System with a Centralized Organization Mahmoud Mofaddel, Djamshid Tavangarian University of Rostock, Department of Computer Science Institut für Technische Informatik Albert-Einstein-Straße

More information

The Effect of Scheduling Discipline on Dynamic Load Sharing in Heterogeneous Distributed Systems

The Effect of Scheduling Discipline on Dynamic Load Sharing in Heterogeneous Distributed Systems Appears in Proc. MASCOTS'97, Haifa, Israel, January 1997. The Effect of Scheduling Discipline on Dynamic Load Sharing in Heterogeneous Distributed Systems Sivarama P. Dandamudi School of Computer Science,

More information

Performance Impact of I/O on Sender-Initiated and Receiver-Initiated Load Sharing Policies in Distributed Systems

Performance Impact of I/O on Sender-Initiated and Receiver-Initiated Load Sharing Policies in Distributed Systems Appears in Proc. Conf. Parallel and Distributed Computing Systems, Dijon, France, 199. Performance Impact of I/O on Sender-Initiated and Receiver-Initiated Load Sharing Policies in Distributed Systems

More information

Study of Load Balancing Schemes over a Video on Demand System

Study of Load Balancing Schemes over a Video on Demand System Study of Load Balancing Schemes over a Video on Demand System Priyank Singhal Ashish Chhabria Nupur Bansal Nataasha Raul Research Scholar, Computer Department Abstract: Load balancing algorithms on Video

More information

NETWORK SIMULATION USING NCTUns. Ankit Verma* Shashi Singh* Meenakshi Vyas*

NETWORK SIMULATION USING NCTUns. Ankit Verma* Shashi Singh* Meenakshi Vyas* NETWORK SIMULATION USING NCTUns Ankit Verma* Shashi Singh* Meenakshi Vyas* 1. Introduction: Network simulator is software which is very helpful tool to develop, test, and diagnose any network protocol.

More information

CS6401- Operating System QUESTION BANK UNIT-I

CS6401- Operating System QUESTION BANK UNIT-I Part-A 1. What is an Operating system? QUESTION BANK UNIT-I An operating system is a program that manages the computer hardware. It also provides a basis for application programs and act as an intermediary

More information

Two Hierarchical Dynamic Load Balancing Algorithms in Distributed Systems

Two Hierarchical Dynamic Load Balancing Algorithms in Distributed Systems 29 Second International Conference on Computer and Electrical Engineering Two Hierarchical Dynamic Load Balancing Algorithms in Distributed Systems Iman Barazandeh Dept. of Computer Engineering IAU, Mahshahr

More information

ANALYSIS OF A DYNAMIC LOAD BALANCING IN MULTIPROCESSOR SYSTEM

ANALYSIS OF A DYNAMIC LOAD BALANCING IN MULTIPROCESSOR SYSTEM International Journal of Computer Science Engineering and Information Technology Research (IJCSEITR) ISSN 2249-6831 Vol. 3, Issue 1, Mar 2013, 143-148 TJPRC Pvt. Ltd. ANALYSIS OF A DYNAMIC LOAD BALANCING

More information

Multimedia-Systems. Operating Systems. Prof. Dr.-Ing. Ralf Steinmetz Prof. Dr. rer. nat. Max Mühlhäuser Prof. Dr.-Ing. Wolfgang Effelsberg

Multimedia-Systems. Operating Systems. Prof. Dr.-Ing. Ralf Steinmetz Prof. Dr. rer. nat. Max Mühlhäuser Prof. Dr.-Ing. Wolfgang Effelsberg Multimedia-Systems Operating Systems Prof. Dr.-Ing. Ralf Steinmetz Prof. Dr. rer. nat. Max Mühlhäuser Prof. Dr.-Ing. Wolfgang Effelsberg WE: University of Mannheim, Dept. of Computer Science Praktische

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

Conflict-free Real-time AGV Routing

Conflict-free Real-time AGV Routing Conflict-free Real-time AGV Routing Rolf H. Möhring, Ekkehard Köhler, Ewgenij Gawrilow, and Björn Stenzel Technische Universität Berlin, Institut für Mathematik, MA 6-1, Straße des 17. Juni 136, 1623 Berlin,

More information

SHARED MEMORY VS DISTRIBUTED MEMORY

SHARED MEMORY VS DISTRIBUTED MEMORY OVERVIEW Important Processor Organizations 3 SHARED MEMORY VS DISTRIBUTED MEMORY Classical parallel algorithms were discussed using the shared memory paradigm. In shared memory parallel platform processors

More information

Job Re-Packing for Enhancing the Performance of Gang Scheduling

Job Re-Packing for Enhancing the Performance of Gang Scheduling Job Re-Packing for Enhancing the Performance of Gang Scheduling B. B. Zhou 1, R. P. Brent 2, C. W. Johnson 3, and D. Walsh 3 1 Computer Sciences Laboratory, Australian National University, Canberra, ACT

More information

Core-Stateless Fair Queueing: Achieving Approximately Fair Bandwidth Allocations in High Speed Networks. Congestion Control in Today s Internet

Core-Stateless Fair Queueing: Achieving Approximately Fair Bandwidth Allocations in High Speed Networks. Congestion Control in Today s Internet Core-Stateless Fair Queueing: Achieving Approximately Fair Bandwidth Allocations in High Speed Networks Ion Stoica CMU Scott Shenker Xerox PARC Hui Zhang CMU Congestion Control in Today s Internet Rely

More information

Improving TCP Performance over Wireless Networks using Loss Predictors

Improving TCP Performance over Wireless Networks using Loss Predictors Improving TCP Performance over Wireless Networks using Loss Predictors Fabio Martignon Dipartimento Elettronica e Informazione Politecnico di Milano P.zza L. Da Vinci 32, 20133 Milano Email: martignon@elet.polimi.it

More information

Technische Universitat Munchen. Institut fur Informatik. D Munchen.

Technische Universitat Munchen. Institut fur Informatik. D Munchen. Developing Applications for Multicomputer Systems on Workstation Clusters Georg Stellner, Arndt Bode, Stefan Lamberts and Thomas Ludwig? Technische Universitat Munchen Institut fur Informatik Lehrstuhl

More information

Improving Http-Server Performance by Adapted Multithreading

Improving Http-Server Performance by Adapted Multithreading Improving Http-Server Performance by Adapted Multithreading Jörg Keller LG Technische Informatik II FernUniversität Hagen 58084 Hagen, Germany email: joerg.keller@fernuni-hagen.de Olaf Monien Thilo Lardon

More information

Chapter 7 CONCLUSION

Chapter 7 CONCLUSION 97 Chapter 7 CONCLUSION 7.1. Introduction A Mobile Ad-hoc Network (MANET) could be considered as network of mobile nodes which communicate with each other without any fixed infrastructure. The nodes in

More information

Lecture 17: Threads and Scheduling. Thursday, 05 Nov 2009

Lecture 17: Threads and Scheduling. Thursday, 05 Nov 2009 CS211: Programming and Operating Systems Lecture 17: Threads and Scheduling Thursday, 05 Nov 2009 CS211 Lecture 17: Threads and Scheduling 1/22 Today 1 Introduction to threads Advantages of threads 2 User

More information

Assignment 5. Georgia Koloniari

Assignment 5. Georgia Koloniari Assignment 5 Georgia Koloniari 2. "Peer-to-Peer Computing" 1. What is the definition of a p2p system given by the authors in sec 1? Compare it with at least one of the definitions surveyed in the last

More information

Single Pass Connected Components Analysis

Single Pass Connected Components Analysis D. G. Bailey, C. T. Johnston, Single Pass Connected Components Analysis, Proceedings of Image and Vision Computing New Zealand 007, pp. 8 87, Hamilton, New Zealand, December 007. Single Pass Connected

More information

Design of Parallel Algorithms. Course Introduction

Design of Parallel Algorithms. Course Introduction + Design of Parallel Algorithms Course Introduction + CSE 4163/6163 Parallel Algorithm Analysis & Design! Course Web Site: http://www.cse.msstate.edu/~luke/courses/fl17/cse4163! Instructor: Ed Luke! Office:

More information

Multiprocessor Scheduling. Multiprocessor Scheduling

Multiprocessor Scheduling. Multiprocessor Scheduling Multiprocessor Scheduling Will consider only shared memory multiprocessor or multi-core CPU Salient features: One or more caches: cache affinity is important Semaphores/locks typically implemented as spin-locks:

More information

Multiprocessor Scheduling

Multiprocessor Scheduling Multiprocessor Scheduling Will consider only shared memory multiprocessor or multi-core CPU Salient features: One or more caches: cache affinity is important Semaphores/locks typically implemented as spin-locks:

More information

6LPXODWLRQÃRIÃWKHÃ&RPPXQLFDWLRQÃ7LPHÃIRUÃDÃ6SDFH7LPH $GDSWLYHÃ3URFHVVLQJÃ$OJRULWKPÃRQÃDÃ3DUDOOHOÃ(PEHGGHG 6\VWHP

6LPXODWLRQÃRIÃWKHÃ&RPPXQLFDWLRQÃ7LPHÃIRUÃDÃ6SDFH7LPH $GDSWLYHÃ3URFHVVLQJÃ$OJRULWKPÃRQÃDÃ3DUDOOHOÃ(PEHGGHG 6\VWHP LPXODWLRQÃRIÃWKHÃ&RPPXQLFDWLRQÃLPHÃIRUÃDÃSDFHLPH $GDSWLYHÃURFHVVLQJÃ$OJRULWKPÃRQÃDÃDUDOOHOÃ(PEHGGHG \VWHP Jack M. West and John K. Antonio Department of Computer Science, P.O. Box, Texas Tech University,

More information

Multiprocessor and Real- Time Scheduling. Chapter 10

Multiprocessor and Real- Time Scheduling. Chapter 10 Multiprocessor and Real- Time Scheduling Chapter 10 Classifications of Multiprocessor Loosely coupled multiprocessor each processor has its own memory and I/O channels Functionally specialized processors

More information

OVERHEADS ENHANCEMENT IN MUTIPLE PROCESSING SYSTEMS BY ANURAG REDDY GANKAT KARTHIK REDDY AKKATI

OVERHEADS ENHANCEMENT IN MUTIPLE PROCESSING SYSTEMS BY ANURAG REDDY GANKAT KARTHIK REDDY AKKATI CMPE 655- MULTIPLE PROCESSOR SYSTEMS OVERHEADS ENHANCEMENT IN MUTIPLE PROCESSING SYSTEMS BY ANURAG REDDY GANKAT KARTHIK REDDY AKKATI What is MULTI PROCESSING?? Multiprocessing is the coordinated processing

More information

Distributed OS and Algorithms

Distributed OS and Algorithms Distributed OS and Algorithms Fundamental concepts OS definition in general: OS is a collection of software modules to an extended machine for the users viewpoint, and it is a resource manager from the

More information

An Efficient Load Balancing Strategy for Scalable WAP Gateways

An Efficient Load Balancing Strategy for Scalable WAP Gateways An Efficient Load Balancing Strategy for Scalable WAP Gateways Te-Hsin Lin and Kuochen Wang Department of Computer and Information Science ational Chiao Tung University Hsinchu,Taiwan 30050,R.O.C. kwang@cis.nctu.edu.tw

More information

Analysis and Algorithms for Partial Protection in Mesh Networks

Analysis and Algorithms for Partial Protection in Mesh Networks Analysis and Algorithms for Partial Protection in Mesh Networks Greg uperman MIT LIDS Cambridge, MA 02139 gregk@mit.edu Eytan Modiano MIT LIDS Cambridge, MA 02139 modiano@mit.edu Aradhana Narula-Tam MIT

More information

Technical Report. Performance Analysis for Parallel R Programs: Towards Efficient Resource Utilization. Helena Kotthaus, Ingo Korb, Peter Marwedel

Technical Report. Performance Analysis for Parallel R Programs: Towards Efficient Resource Utilization. Helena Kotthaus, Ingo Korb, Peter Marwedel Performance Analysis for Parallel R Programs: Towards Efficient Resource Utilization Technical Report Helena Kotthaus, Ingo Korb, Peter Marwedel 01/2015 technische universität dortmund Part of the work

More information

CHAPTER 3 EFFECTIVE ADMISSION CONTROL MECHANISM IN WIRELESS MESH NETWORKS

CHAPTER 3 EFFECTIVE ADMISSION CONTROL MECHANISM IN WIRELESS MESH NETWORKS 28 CHAPTER 3 EFFECTIVE ADMISSION CONTROL MECHANISM IN WIRELESS MESH NETWORKS Introduction Measurement-based scheme, that constantly monitors the network, will incorporate the current network state in the

More information

The MOSIX Algorithms for Managing Cluster, Multi-Clusters, GPU Clusters and Clouds

The MOSIX Algorithms for Managing Cluster, Multi-Clusters, GPU Clusters and Clouds The MOSIX Algorithms for Managing Cluster, Multi-Clusters, GPU Clusters and Clouds Prof. Amnon Barak Department of Computer Science The Hebrew University of Jerusalem http:// www. MOSIX. Org 1 Background

More information

Performance Evaluation of Mesh - Based Multicast Routing Protocols in MANET s

Performance Evaluation of Mesh - Based Multicast Routing Protocols in MANET s Performance Evaluation of Mesh - Based Multicast Routing Protocols in MANET s M. Nagaratna Assistant Professor Dept. of CSE JNTUH, Hyderabad, India V. Kamakshi Prasad Prof & Additional Cont. of. Examinations

More information

LECTURE 3:CPU SCHEDULING

LECTURE 3:CPU SCHEDULING LECTURE 3:CPU SCHEDULING 1 Outline Basic Concepts Scheduling Criteria Scheduling Algorithms Multiple-Processor Scheduling Real-Time CPU Scheduling Operating Systems Examples Algorithm Evaluation 2 Objectives

More information

Load Balancing for Problems with Good Bisectors, and Applications in Finite Element Simulations

Load Balancing for Problems with Good Bisectors, and Applications in Finite Element Simulations Load Balancing for Problems with Good Bisectors, and Applications in Finite Element Simulations Stefan Bischof, Ralf Ebner, and Thomas Erlebach Institut für Informatik Technische Universität München D-80290

More information

Lecture Topics. Announcements. Today: Advanced Scheduling (Stallings, chapter ) Next: Deadlock (Stallings, chapter

Lecture Topics. Announcements. Today: Advanced Scheduling (Stallings, chapter ) Next: Deadlock (Stallings, chapter Lecture Topics Today: Advanced Scheduling (Stallings, chapter 10.1-10.4) Next: Deadlock (Stallings, chapter 6.1-6.6) 1 Announcements Exam #2 returned today Self-Study Exercise #10 Project #8 (due 11/16)

More information

Scheduling in the Supermarket

Scheduling in the Supermarket Scheduling in the Supermarket Consider a line of people waiting in front of the checkout in the grocery store. In what order should the cashier process their purchases? Scheduling Criteria CPU utilization

More information

Connectivity, Energy and Mobility Driven Clustering Algorithm for Mobile Ad Hoc Networks

Connectivity, Energy and Mobility Driven Clustering Algorithm for Mobile Ad Hoc Networks Connectivity, Energy and Mobility Driven Clustering Algorithm for Mobile Ad Hoc Networks Fatiha Djemili Tolba University of Haute Alsace GRTC Colmar, France fatiha.tolba@uha.fr Damien Magoni University

More information

Chapter 11: Implementing File-Systems

Chapter 11: Implementing File-Systems Chapter 11: Implementing File-Systems Chapter 11 File-System Implementation 11.1 File-System Structure 11.2 File-System Implementation 11.3 Directory Implementation 11.4 Allocation Methods 11.5 Free-Space

More information

Multiprocessor and Real-Time Scheduling. Chapter 10

Multiprocessor and Real-Time Scheduling. Chapter 10 Multiprocessor and Real-Time Scheduling Chapter 10 1 Roadmap Multiprocessor Scheduling Real-Time Scheduling Linux Scheduling Unix SVR4 Scheduling Windows Scheduling Classifications of Multiprocessor Systems

More information

Basic Switch Organization

Basic Switch Organization NOC Routing 1 Basic Switch Organization 2 Basic Switch Organization Link Controller Used for coordinating the flow of messages across the physical link of two adjacent switches 3 Basic Switch Organization

More information

Clustering-Based Distributed Precomputation for Quality-of-Service Routing*

Clustering-Based Distributed Precomputation for Quality-of-Service Routing* Clustering-Based Distributed Precomputation for Quality-of-Service Routing* Yong Cui and Jianping Wu Department of Computer Science, Tsinghua University, Beijing, P.R.China, 100084 cy@csnet1.cs.tsinghua.edu.cn,

More information

Load Balancing with Random Information Exchanged based Policy

Load Balancing with Random Information Exchanged based Policy Load Balancing with Random Information Exchanged based Policy Taj Alam 1, Zahid Raza 2 School of Computer & Systems Sciences Jawaharlal Nehru University New Delhi, India 1 tajhashimi@gmail.com, 2 zahidraza@mail.jnu.ac.in

More information

An Improved Priority Dynamic Quantum Time Round-Robin Scheduling Algorithm

An Improved Priority Dynamic Quantum Time Round-Robin Scheduling Algorithm An Improved Priority Dynamic Quantum Time Round-Robin Scheduling Algorithm Nirali A. Patel PG Student, Information Technology, L.D. College Of Engineering,Ahmedabad,India ABSTRACT In real-time embedded

More information

Distributed Systems 27. Process Migration & Allocation

Distributed Systems 27. Process Migration & Allocation Distributed Systems 27. Process Migration & Allocation Paul Krzyzanowski pxk@cs.rutgers.edu 12/16/2011 1 Processor allocation Easy with multiprocessor systems Every processor has access to the same memory

More information

Fault-Tolerant Dynamic Task Scheduling Based on Dataflow Graphs *

Fault-Tolerant Dynamic Task Scheduling Based on Dataflow Graphs * - 1 - Fault-Tolerant Dynamic Task Scheduling Based on Dataflow Graphs * Erik Maehle and Franz-Josef Markus Medizinische Universität zu Lübeck, Institut für Technische Informatik Ratzeburger Allee 16, D-23538

More information

Course Syllabus. Operating Systems

Course Syllabus. Operating Systems Course Syllabus. Introduction - History; Views; Concepts; Structure 2. Process Management - Processes; State + Resources; Threads; Unix implementation of Processes 3. Scheduling Paradigms; Unix; Modeling

More information

Performance Evaluation of Mobile Agent-based Dynamic Load Balancing Algorithm

Performance Evaluation of Mobile Agent-based Dynamic Load Balancing Algorithm Performance Evaluation of Mobile -based Dynamic Load Balancing Algorithm MAGDY SAEB, CHERINE FATHY Computer Engineering Department Arab Academy for Science, Technology & Maritime Transport Alexandria,

More information

A priority based dynamic bandwidth scheduling in SDN networks 1

A priority based dynamic bandwidth scheduling in SDN networks 1 Acta Technica 62 No. 2A/2017, 445 454 c 2017 Institute of Thermomechanics CAS, v.v.i. A priority based dynamic bandwidth scheduling in SDN networks 1 Zun Wang 2 Abstract. In order to solve the problems

More information

Web page recommendation using a stochastic process model

Web page recommendation using a stochastic process model Data Mining VII: Data, Text and Web Mining and their Business Applications 233 Web page recommendation using a stochastic process model B. J. Park 1, W. Choi 1 & S. H. Noh 2 1 Computer Science Department,

More information

Multimedia Systems 2011/2012

Multimedia Systems 2011/2012 Multimedia Systems 2011/2012 System Architecture Prof. Dr. Paul Müller University of Kaiserslautern Department of Computer Science Integrated Communication Systems ICSY http://www.icsy.de Sitemap 2 Hardware

More information

Performance Study of Routing Algorithms for LEO Satellite Constellations

Performance Study of Routing Algorithms for LEO Satellite Constellations Performance Study of Routing Algorithms for LEO Satellite Constellations Ioannis Gragopoulos, Evangelos Papapetrou, Fotini-Niovi Pavlidou Aristotle University of Thessaloniki, School of Engineering Dept.

More information

Making Workstations a Friendly Environment for Batch Jobs. Miron Livny Mike Litzkow

Making Workstations a Friendly Environment for Batch Jobs. Miron Livny Mike Litzkow Making Workstations a Friendly Environment for Batch Jobs Miron Livny Mike Litzkow Computer Sciences Department University of Wisconsin - Madison {miron,mike}@cs.wisc.edu 1. Introduction As time-sharing

More information

File-System Structure

File-System Structure Chapter 12: File System Implementation File System Structure File System Implementation Directory Implementation Allocation Methods Free-Space Management Efficiency and Performance Recovery Log-Structured

More information

A Comparison of Two Distributed Systems: Amoeba & Sprite. By: Fred Douglis, John K. Ousterhout, M. Frans Kaashock, Andrew Tanenbaum Dec.

A Comparison of Two Distributed Systems: Amoeba & Sprite. By: Fred Douglis, John K. Ousterhout, M. Frans Kaashock, Andrew Tanenbaum Dec. A Comparison of Two Distributed Systems: Amoeba & Sprite By: Fred Douglis, John K. Ousterhout, M. Frans Kaashock, Andrew Tanenbaum Dec. 1991 Introduction shift from time-sharing to multiple processors

More information

Distributed STDMA in Ad Hoc Networks

Distributed STDMA in Ad Hoc Networks Distributed STDMA in Ad Hoc Networks Jimmi Grönkvist Swedish Defence Research Agency SE-581 11 Linköping, Sweden email: jimgro@foi.se Abstract Spatial reuse TDMA is a collision-free access scheme for ad

More information

Live Virtual Machine Migration with Efficient Working Set Prediction

Live Virtual Machine Migration with Efficient Working Set Prediction 2011 International Conference on Network and Electronics Engineering IPCSIT vol.11 (2011) (2011) IACSIT Press, Singapore Live Virtual Machine Migration with Efficient Working Set Prediction Ei Phyu Zaw

More information

SMD149 - Operating Systems - Multiprocessing

SMD149 - Operating Systems - Multiprocessing SMD149 - Operating Systems - Multiprocessing Roland Parviainen December 1, 2005 1 / 55 Overview Introduction Multiprocessor systems Multiprocessor, operating system and memory organizations 2 / 55 Introduction

More information

Overview. SMD149 - Operating Systems - Multiprocessing. Multiprocessing architecture. Introduction SISD. Flynn s taxonomy

Overview. SMD149 - Operating Systems - Multiprocessing. Multiprocessing architecture. Introduction SISD. Flynn s taxonomy Overview SMD149 - Operating Systems - Multiprocessing Roland Parviainen Multiprocessor systems Multiprocessor, operating system and memory organizations December 1, 2005 1/55 2/55 Multiprocessor system

More information

Performance of Multihop Communications Using Logical Topologies on Optical Torus Networks

Performance of Multihop Communications Using Logical Topologies on Optical Torus Networks Performance of Multihop Communications Using Logical Topologies on Optical Torus Networks X. Yuan, R. Melhem and R. Gupta Department of Computer Science University of Pittsburgh Pittsburgh, PA 156 fxyuan,

More information

Lecture: Interconnection Networks

Lecture: Interconnection Networks Lecture: Interconnection Networks Topics: Router microarchitecture, topologies Final exam next Tuesday: same rules as the first midterm 1 Packets/Flits A message is broken into multiple packets (each packet

More information

ALL the assignments (A1, A2, A3) and Projects (P0, P1, P2) we have done so far.

ALL the assignments (A1, A2, A3) and Projects (P0, P1, P2) we have done so far. Midterm Exam Reviews ALL the assignments (A1, A2, A3) and Projects (P0, P1, P2) we have done so far. Particular attentions on the following: System call, system kernel Thread/process, thread vs process

More information

CS370 Operating Systems

CS370 Operating Systems CS370 Operating Systems Colorado State University Yashwant K Malaiya Fall 2017 Lecture 10 Slides based on Text by Silberschatz, Galvin, Gagne Various sources 1 1 Chapter 6: CPU Scheduling Basic Concepts

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

Department of Information Technology Sri Venkateshwara College of Engineering, Chennai, India. 1 2

Department of Information Technology Sri Venkateshwara College of Engineering, Chennai, India. 1 2 Energy-Aware Scheduling Using Workload Consolidation Techniques in Cloud Environment 1 Sridharshini V, 2 V.M.Sivagami 1 PG Scholar, 2 Associate Professor Department of Information Technology Sri Venkateshwara

More information

Dynamic Routing and Resource Allocation in WDM Transport Networks

Dynamic Routing and Resource Allocation in WDM Transport Networks Dynamic Routing and Resource Allocation in WDM Transport Networks Jan Späth University of Stuttgart, Institute of Communication Networks and Computer Engineering (IND), Germany Email: spaeth@ind.uni-stuttgart.de

More information

Lecture 12: Interconnection Networks. Topics: communication latency, centralized and decentralized switches, routing, deadlocks (Appendix E)

Lecture 12: Interconnection Networks. Topics: communication latency, centralized and decentralized switches, routing, deadlocks (Appendix E) Lecture 12: Interconnection Networks Topics: communication latency, centralized and decentralized switches, routing, deadlocks (Appendix E) 1 Topologies Internet topologies are not very regular they grew

More information

Dynamic Load balancing for I/O- and Memory- Intensive workload in Clusters using a Feedback Control Mechanism

Dynamic Load balancing for I/O- and Memory- Intensive workload in Clusters using a Feedback Control Mechanism Dynamic Load balancing for I/O- and Memory- Intensive workload in Clusters using a Feedback Control Mechanism Xiao Qin, Hong Jiang, Yifeng Zhu, David R. Swanson Department of Computer Science and Engineering

More information

Motivation. Threads. Multithreaded Server Architecture. Thread of execution. Chapter 4

Motivation. Threads. Multithreaded Server Architecture. Thread of execution. Chapter 4 Motivation Threads Chapter 4 Most modern applications are multithreaded Threads run within application Multiple tasks with the application can be implemented by separate Update display Fetch data Spell

More information

Distributed Clustering Method for Large-Scaled Wavelength Routed Networks

Distributed Clustering Method for Large-Scaled Wavelength Routed Networks Distributed Clustering Method for Large-Scaled Wavelength Routed Networks Yukinobu Fukushima Graduate School of Information Science and Technology, Osaka University - Yamadaoka, Suita, Osaka 60-08, Japan

More information

Random Early Detection (RED) gateways. Sally Floyd CS 268: Computer Networks

Random Early Detection (RED) gateways. Sally Floyd CS 268: Computer Networks Random Early Detection (RED) gateways Sally Floyd CS 268: Computer Networks floyd@eelblgov March 20, 1995 1 The Environment Feedback-based transport protocols (eg, TCP) Problems with current Drop-Tail

More information

C3PO: Computation Congestion Control (PrOactive)

C3PO: Computation Congestion Control (PrOactive) C3PO: Computation Congestion Control (PrOactive) an algorithm for dynamic diffusion of ephemeral in-network services Liang Wang, Mario Almeida*, Jeremy Blackburn*, Jon Crowcroft University of Cambridge,

More information

Scheduling. Scheduling 1/51

Scheduling. Scheduling 1/51 Scheduling 1/51 Scheduler Scheduling Scheduler allocates cpu(s) to threads and processes. This action is known as scheduling. The scheduler is a part of the process manager code that handles scheduling.

More information

Optimized DTN-Routing for Urban Public Transport Systems

Optimized DTN-Routing for Urban Public Transport Systems Optimized DTN-Routing for Urban Public Transport Systems Tobias Pögel Institute of Operating Systems and Computer Networks Technische Universität Braunschweig, Germany poegel@ibr.cs.tu-bs.de Abstract Communication

More information

High Performance Computing. University questions with solution

High Performance Computing. University questions with solution High Performance Computing University questions with solution Q1) Explain the basic working principle of VLIW processor. (6 marks) The following points are basic working principle of VLIW processor. The

More information

From Cluster Monitoring to Grid Monitoring Based on GRM *

From Cluster Monitoring to Grid Monitoring Based on GRM * From Cluster Monitoring to Grid Monitoring Based on GRM * Zoltán Balaton, Péter Kacsuk, Norbert Podhorszki and Ferenc Vajda MTA SZTAKI H-1518 Budapest, P.O.Box 63. Hungary {balaton, kacsuk, pnorbert, vajda}@sztaki.hu

More information

Navjot Jyoti ABSTRACT I. INTRODUCTION

Navjot Jyoti ABSTRACT I. INTRODUCTION International Journal of Scientific esearch in Computer Science, Engineering and Information echnology 217 IJSCSEI Volume 2 Issue 1 ISSN : 2456-337 An Analytical eview : Static Load Balancing Algorithms

More information

OPERATING SYSTEM. Chapter 12: File System Implementation

OPERATING SYSTEM. Chapter 12: File System Implementation OPERATING SYSTEM Chapter 12: File System Implementation Chapter 12: File System Implementation File-System Structure File-System Implementation Directory Implementation Allocation Methods Free-Space Management

More information

DISTRIBUTED HIGH-SPEED COMPUTING OF MULTIMEDIA DATA

DISTRIBUTED HIGH-SPEED COMPUTING OF MULTIMEDIA DATA DISTRIBUTED HIGH-SPEED COMPUTING OF MULTIMEDIA DATA M. GAUS, G. R. JOUBERT, O. KAO, S. RIEDEL AND S. STAPEL Technical University of Clausthal, Department of Computer Science Julius-Albert-Str. 4, 38678

More information

An Efficient Live Process Migration Approach for High Performance Cluster Computing Systems

An Efficient Live Process Migration Approach for High Performance Cluster Computing Systems An Efficient Live Process Migration Approach for High Performance Cluster Computing Systems Ehsan Mousavi Khaneghah, Najmeh Osouli Nezhad, Seyedeh Leili Mirtaheri, Mohsen Sharifi, and Ashakan Shirpour

More information

New Forwarding Strategy for Metro Ethernet Networks Based on Hierarchical Addressing

New Forwarding Strategy for Metro Ethernet Networks Based on Hierarchical Addressing New Forwarding Strategy for Metro Ethernet Networks Based on Hierarchical Addressing Farhad Faghani*, Ghasem Mirjalily**, Reza Saadat**, Farmarz Hendessi*** *Department of Electrical Engineering, Najafabad

More information

Process- Concept &Process Scheduling OPERATING SYSTEMS

Process- Concept &Process Scheduling OPERATING SYSTEMS OPERATING SYSTEMS Prescribed Text Book Operating System Principles, Seventh Edition By Abraham Silberschatz, Peter Baer Galvin and Greg Gagne PROCESS MANAGEMENT Current day computer systems allow multiple

More information

Coarse-to-Fine Search Technique to Detect Circles in Images

Coarse-to-Fine Search Technique to Detect Circles in Images Int J Adv Manuf Technol (1999) 15:96 102 1999 Springer-Verlag London Limited Coarse-to-Fine Search Technique to Detect Circles in Images M. Atiquzzaman Department of Electrical and Computer Engineering,

More information

Chapter 6: CPU Scheduling. Operating System Concepts 9 th Edition

Chapter 6: CPU Scheduling. Operating System Concepts 9 th Edition Chapter 6: CPU Scheduling Silberschatz, Galvin and Gagne 2013 Chapter 6: CPU Scheduling Basic Concepts Scheduling Criteria Scheduling Algorithms Thread Scheduling Multiple-Processor Scheduling Real-Time

More information

A Modified Inertial Method for Loop-free Decomposition of Acyclic Directed Graphs

A Modified Inertial Method for Loop-free Decomposition of Acyclic Directed Graphs MACRo 2015-5 th International Conference on Recent Achievements in Mechatronics, Automation, Computer Science and Robotics A Modified Inertial Method for Loop-free Decomposition of Acyclic Directed Graphs

More information

Abstract A SCALABLE, PARALLEL, AND RECONFIGURABLE DATAPATH ARCHITECTURE

Abstract A SCALABLE, PARALLEL, AND RECONFIGURABLE DATAPATH ARCHITECTURE A SCALABLE, PARALLEL, AND RECONFIGURABLE DATAPATH ARCHITECTURE Reiner W. Hartenstein, Rainer Kress, Helmut Reinig University of Kaiserslautern Erwin-Schrödinger-Straße, D-67663 Kaiserslautern, Germany

More information

CPU THREAD PRIORITIZATION USING A DYNAMIC QUANTUM TIME ROUND-ROBIN ALGORITHM

CPU THREAD PRIORITIZATION USING A DYNAMIC QUANTUM TIME ROUND-ROBIN ALGORITHM CPU THREAD PRIORITIZATION USING A DYNAMIC QUANTUM TIME ROUND-ROBIN ALGORITHM Maysoon A. Mohammed 1, 2, Mazlina Abdul Majid 1, Balsam A. Mustafa 1 and Rana Fareed Ghani 3 1 Faculty of Computer System &

More information

PROXIMITY AWARE LOAD BALANCING FOR HETEROGENEOUS NODES Mrs. Yogita A. Dalvi Dr. R. Shankar Mr. Atesh Kumar

PROXIMITY AWARE LOAD BALANCING FOR HETEROGENEOUS NODES Mrs. Yogita A. Dalvi Dr. R. Shankar Mr. Atesh Kumar ISSN 2320-9194 1 International Journal of Advance Research, IJOAR.org Volume 1, Issue 9, September 2013, Online: ISSN 2320-9194 PROXIMITY AWARE LOAD BALANCING FOR HETEROGENEOUS NODES Mrs. Yogita A. Dalvi

More information

The MOSIX Scalable Cluster Computing for Linux. mosix.org

The MOSIX Scalable Cluster Computing for Linux.  mosix.org The MOSIX Scalable Cluster Computing for Linux Prof. Amnon Barak Computer Science Hebrew University http://www. mosix.org 1 Presentation overview Part I : Why computing clusters (slide 3-7) Part II : What

More information

Video Inter-frame Forgery Identification Based on Optical Flow Consistency

Video Inter-frame Forgery Identification Based on Optical Flow Consistency Sensors & Transducers 24 by IFSA Publishing, S. L. http://www.sensorsportal.com Video Inter-frame Forgery Identification Based on Optical Flow Consistency Qi Wang, Zhaohong Li, Zhenzhen Zhang, Qinglong

More information

CHAPTER 2: PROCESS MANAGEMENT

CHAPTER 2: PROCESS MANAGEMENT 1 CHAPTER 2: PROCESS MANAGEMENT Slides by: Ms. Shree Jaswal TOPICS TO BE COVERED Process description: Process, Process States, Process Control Block (PCB), Threads, Thread management. Process Scheduling:

More information

Multiprocessor Scheduling. Multiprocessor Scheduling

Multiprocessor Scheduling. Multiprocessor Scheduling Multiprocessor Scheduling Will consider only shared memory multiprocessor or multi-core CPU Salient features: One or more caches: cache affinity is important Semaphores/locks typically implemented as spin-locks:

More information

Verification and Validation of X-Sim: A Trace-Based Simulator

Verification and Validation of X-Sim: A Trace-Based Simulator http://www.cse.wustl.edu/~jain/cse567-06/ftp/xsim/index.html 1 of 11 Verification and Validation of X-Sim: A Trace-Based Simulator Saurabh Gayen, sg3@wustl.edu Abstract X-Sim is a trace-based simulator

More information

What is Parallel Computing?

What is Parallel Computing? What is Parallel Computing? Parallel Computing is several processing elements working simultaneously to solve a problem faster. 1/33 What is Parallel Computing? Parallel Computing is several processing

More information

Chapter 17: Distributed Systems (DS)

Chapter 17: Distributed Systems (DS) Chapter 17: Distributed Systems (DS) Silberschatz, Galvin and Gagne 2013 Chapter 17: Distributed Systems Advantages of Distributed Systems Types of Network-Based Operating Systems Network Structure Communication

More information

Operating System Performance and Large Servers 1

Operating System Performance and Large Servers 1 Operating System Performance and Large Servers 1 Hyuck Yoo and Keng-Tai Ko Sun Microsystems, Inc. Mountain View, CA 94043 Abstract Servers are an essential part of today's computing environments. High

More information

CPU Scheduling: Objectives

CPU Scheduling: Objectives CPU Scheduling: Objectives CPU scheduling, the basis for multiprogrammed operating systems CPU-scheduling algorithms Evaluation criteria for selecting a CPU-scheduling algorithm for a particular system

More information