Thesis (M.C.S.) - Carleton University, 2007. / Includes bibliographical references (p. 74-77). Also available in electronic format on the Internet.

Thesis (MEngSc)--University of Melbourne, Faculty of Engineering, 2010. / Typescript. Includes bibliographical references (p. 70-74)

Thesis (M.S.)--Brigham Young University. Dept. of Computer Science, 2005. / Includes bibliographical references (p. 59-60).

Thesis (Ph.D.)--University of Nebraska-Lincoln, 2006. / Title from title screen (site viewed on January 23, 2007). PDF text: viii, 137 p. : ill. ; 0.85Mb. UMI publication number: AAT 3215322. Includes bibliographical references. Also available in microfilm and microfiche format.

Thesis (Ph.D.)--University of Nebraska-Lincoln, 2007. / Title from title screen (site viewed Oct. 10, 2007). PDF text: 135 p. : ill. (some col.) UMI publication number: AAT 3258771. Includes bibliographical references. Also available in microfilm and microfiche formats.

With the advent in multicore computers, the scheduling of Grid jobs can be made more effective if scaled to fully utilize the underlying hardware and parallelized to benefit from the exploitation of multicores. The fact that sequential algorithms do not scale with multicore systems nor benefit from parallelism remains a major challenge to scheduling in the Grid. As multicore systems become ever more pervasive in our computing lives, over reliance on such systems for passive parallelism does not offer the best option in harnessing the benefits of their multiprocessors for Grid scheduling. An explicit means of exploiting parallelism for Grid scheduling is required. The Group-based Parallel Multi-scheduler for Grid introduced in this work is aimed at effectively exploiting the benefits of multicore systems for Grid job scheduling by splitting jobs and machines into paired groups and independently multi-scheduling jobs in parallel from the groups. The Priority method splits jobs into four priority groups based on job attributes and uses two methods (SimTog and EvenDist) methods to group machines. Then the scheduling is carried out using the MinMin algorithm within the discrete group pairs. The Priority method was implemented and compared with the MinMin scheduling algorithm without grouping (named ordinary MinMin in this research). The analysis of results compared against the ordinary MinMin shows substantial improvement in speedup and gains in scheduling efficiency. In addition, the Execution Time Balanced (ETB) and Execution Time Sorted then Balanced (ETSB) methods were also implemented to group jobs in order to improve on some deficiencies found with the Priority method. The two methods used the same machine grouping methods as used with the Priority method, but were able to vary the number of groups and equally exploited different means of grouping jobs to ensure equitability of jobs in groups. The MinMin Grid scheduling algorithm was then executed independently within the discrete group pairs. Results and analysis shows that the ETB and ETSB methods gain still further improvement over MinMin compared to the Priority method. The conclusion is reached that grouping jobs and machines before scheduling improves the scheduling efficiency significantly.

Thesis (MPhil)--Stellenbosch University, 2005. / ENGLISH ABSTRACT: Prior to the invention of Personal Computers the scope of research activities was limited by the pre-existing capabilities of problem solving mechanisms. However, with the advent of PCs and inter-networking thereof, the new tools (hardware and software) enabled the scientific community to tackle more complex research challenges and this led to a better understanding of our environment. The development of the Internet also enabled research communities to communicate and share information in real time. However, even the Internet has limitations of its own when it comes to the need of sharing not only information but also massive storage, processing power, huge databases and applications, expensive and delicate scientific instruments, knowledge and expertise. This led to the need for a networking system that includes these above-mentioned services, using the Internet infrastructure, semantic web technologies and pervasive computing devices, which is so called Grid Computing. This research study deals with a Generic Campus Grid Computing framework, which mobilizes the available idle/extra computing resources residing in the faculty-computing centres for use by the e-community on CPU-intensive or Data-intensive jobs. This unused computing capacity could be utilized for Grid computing services; hence, the already available resources could be more efficiently exploited. Besides, this could be a huge saving when compared to the cost of acquiring supercomputers by these institutions. Therefore, this research study intends to establish a simple and functional Generic Campus Grid Computing Framework at this stage, with the consent that subsequent research studies could deal with further assessment in a more detailed perspective and practical implementation thereof. / AFRIKAANSE OPSOMMING: Voor die uitvinding van die Persoonlike Rekenaar is die omvang van navorsingsaktiwiteite beperk deur die voorafbestaande vermoëns van probleemoplossingsmeganismes. Met die verskyning van PR's en die daaropvolgende internetwerking daarvan, het die nuwe gereedskap (hardeware en sagteware) die wetenskaplike gemeenskap in staat gestel om meer komplekse navorsingsuitdagings aan te pak. Dit het gelei tot groter begrip van ons omgewing. Die onwikkeling van die Internet het navorsingsgemeenskappe ook in staat gestel om in reële tyd te kommunikeer en inligting te deel. Nietemin, selfs die Internet het gebreke wanneer dit kom by die behoefte om nie slegs inligting te deel nie, maar ook massiewe stoorruimte, verwerkingskrag, baie groot databasisse en toepassings, duur en delikate wetenskaplike toerusting, kennis en kundigheid. Dit het gelei tot die behoefte aan 'n netwerksisteem wat bogenoemde dienste insluit, deur gebruik te maak van Internet-infrastruktuur, semantiese web tegnologieë, en alomteenwoordige rekenaartoestelle. Hierdie sisteem staan bekend as "Grid Computing" of te wel Rooster Komputasie. Hierdie navorsingstudie handel oor 'n Generiese Kampus Rooster Komputasie Raamwerk wat die ongebruikte, ekstra komputasiebronne, wat beskikbaar is in fakulteite se rekenaargebruikersareas, mobiliseer vir gebruik deur die e-gemeenskap op SVE-intensiewe of Dataintensiewe toepassings. Hierdie ongebruikte komputasie kapasiteit kan aangewend word vir Rooster komputasie dienste; gevolglik kan die beskikbare bronne dan meer effektief benut word. Verder kan dit lei tot groot besparings wanneer dit vergelyk word met die koste om superrekenaars aan te koop deur die betrokke instansies. Dus, op hierdie stadium stel hierdie navorsingstudie dit ten doel om 'n eenvoudige en funksionele Generiese Kampus Rooster Komputasie Raamwerk te skep met dien verstande dat daaropvolgende studies sou kon fokus op verdere assessering met 'n meer gedetaileerde perspektief en met praktiese implementasie.

The rise of business-oriented and commercial applications for Grid computing environments has recently gathered pace. Grid computing traditionally has been linked with scientific environments, where heterogeneous resources provided by Grid systems and infrastructures were employed for carrying out computationally-intensive and data-intensive scientific experiments or applications that may have not been possible before. The natural progression is that business-oriented applications will look to build on this success and utilise the large number of heterogeneous Grid resources including computational resources such as CPUs and memory and storage resources such as disk space, potentially available. The success of introducing these applications into the mainstream is directly related to whether service providers can deliver a level of Quality of Service (QoS) to a consumer and the ability of the consumer to request high-level QoS such as the numbers of CPUs required or the RAM required. QoS refers to the guidelines and requirements requested by a user/consumer from the service providers and resources. The communication and agreement establishment processes between user and provider must be defined clearly to accommodate a new type of user where knowledge of the underlying infrastructure cannot be assumed. QoS parameters have generally been defined at the Grid resource level using low level definitions. This tailors to specific applications and models related to scientific domains where brokering, scheduling and QoS delivery is designed for specific applications within specific domains. This thesis presents a flexible model for high-level QoS requests. Business Grid Quality of Service (BGQoS) is introduced for business-oriented and commercial Grid applications which may wish to make use of the resources made available by Grid system environments. BGQoS allows GRCs (Grid Resource Consumers) to specify varying types of high-level QoS requirements which are delivered via querying up-to-date resource information, matchmaking and monitoring operations. Moreover, we present dynamically calculated metrics for measuring QoS such as reliability, increasing the accuracy of meeting the GRC’s requirements. On the other hand GRPs (Grid Resource Provider) are also capable of advertising their resources, their capabilities, their usage policies and availability both locally and globally. This leads to a flexible model that could be carried across domains without altering the core operations and which could easily be expanded in order to accommodate different types of GRC, resources and applications.

published_or_final_version / abstract / Computer Science and Information Systems / Master / Master of Philosophy

M.Sc. / Grid computing is a technology concerned with harvesting idle resources of geographically distributed and interconnected computers. It solves problems regarded as too complex or large to be solved by a single computer. Furthermore, economic grid computing is becoming the most dominant form of grid computing. It enables some form of payment to occur between resource producers and resource consumers in grid computing. Mobile devices and mobile telecommunication services, a relatively new field of technology, are rapidly increasing in popularity, size, strength and application. At the end of 2006, there were approximately 2.7 billion global active mobile users utilising mobile devices and mobile telecommunication services [Aho07]. At the end of 2007 this number had grown to 3.3 billion mobile users, more than half a billion additional mobile users in a period of one year [McN07]. With such large numbers, grid computing can benefit from the clustering of mobile devices forming a mobile grid computing model. However, there are many inherent disadvantages concerning mobile devices, such as low processing capabilities, unpredictable network connections and battery utilisation. Such hurdles must be addressed and solved if a mobile computing infrastructure or architecture is ever to be considered. This dissertation proposes the implementation of an economic mobile computing solution: Mobile Agents for Global Mobile Device Grid Infrastructure Enterprises, or MAGGIE. MAGGIE is concerned with harvesting idle mobile device resources by implementing the supply and demand economic model, aiming to create a healthy competitive economic market environment. MAGGIE implements agent and mobile agent technology to compensate for the hurdles introduced by mobile devices and mobile device software development platforms. It is targeted at both Sun Microsystems’s J2ME MIDP 2.0 and Microsoft’s .NET Compact Framework, enabling lower-end and higher-end mobile devices to contribute mobile computing services and resources for utilisation by other mobile device users. The primary goal of MAGGIE is to produce an architecture as generic as possible regarding the development and implementation of MAGGIE services. MAGGIE allows third-party application developers to seamlessly implement an array of MAGGIE services, without indepth prior knowledge of the architecture and technical aspects of MAGGIE. Finally, MAGGIE’s capabilities are demonstrated by implementing a distributed mobile chess service known as the MAGGIE Chess Service. The MAGGIE Chess Service enables a collection of distributed mobile devices in determining the best move originating from a chessboard position.