I. Foster and C. Kesselman, (eds.). Computational Grids: The Future of High Performance Distributed Computing. Morgan Kaufman, San Mateo, CA., 1998.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....scalability at all. Other multi agent frameworks rely on another framework to solve the problem of scalability. For example, scalablity in the CoABS (DARPA Control of Agent Based Systems) approach [32] is based adequate support from computational grids in providing a plug in backplane for agents [11]. In other multi agent frameworks, aspects of scalability are specifically addressed. In ZEUS [38] scalability is defined to be the growth rate of the maximum communication load (as a function of the number of agents) Their conclusions are that the maximum communication load grows at worst ....

I. Foster and C. Kesselman, (eds.). Computational Grids: The Future of High Performance Distributed Computing. Morgan Kaufman, San Mateo, CA., 1998.

....at all. Other multi agent frameworks rely on another framework to solve the problem of scalability. For example, scalablity in the CoABS (DARPA Control of Agent Based Systems) approach [27] is based on adequate support from computational grids in providing a plug in backplane for agents [8]. There are, however, frameworks that clearly address one or more aspects of scalability. In ZEUS [29] scalability is defined to be the growth rate of the maximum communication load (as a function of the number of agents) Their conclusions are that the maximum communication load grows at worst ....

I. Foster and C. Kesselman, (eds.). Computational Grids: The Future of High Performance Distributed Computing. Morgan Kaufman, San Mateo, CA., 1998.

....is an issue. Some multi agent frameworks rely on another framework to solve the problem of scalability. For example, scalability in the CoABS (DARPA Control of Agent Based Systems) approach [55] is based on adequate support from computational grids in providing a plug in infrastructure for agents [19]. 7 In other multi agent frameworks, aspects of scalability are specifically addressed. In ZEUS [13] scalability is defined to be the growth rate of the maximum communication load, that is, the number of messages sent across a single link, grows at worst linearly with the number of agents. This ....

I. Foster and C. Kesselman, editors. Computational Grids: The Future of High Performance Distributed Computing. Morgan Kaufman, San Mateo, CA, 1998. 17

....scalability at all. Other multi agent frameworks rely on another framework to solve the problem of scalability. For example, scalablity in the CoABS (DARPA Control of Agent Based Systems) approach [32] is based adequate support from computational grids in providing a plug in backplane for agents [11]. In other multi agent frameworks, aspects of scalability are speci cally addressed. In ZEUS [38] scalability is de ned to be the growth rate of the maximum communication load (as a function of the number of agents) Their conclusions are that the maximum communication load grows at worst ....

I. Foster and C. Kesselman, (eds.). Computational Grids: The Future of High Performance Distributed Computing. Morgan Kaufman, San Mateo, CA., 1998.

....at all. Other multi agent frameworks rely on another framework to solve the problem of scalability. For example, scalablity in the CoABS (DARPA Control of Agent Based Systems) approach [29] is based on adequate support from computational grids in providing a plug in backplane for agents [8]. There are, however, frameworks that clearly address one or more aspects of scalability. In ZEUS [31] scalability is de ned to be the growth rate of the maximum communication load (as a function of the number of agents) Their conclusions are that the maximum communication load grows at worst ....

I. Foster and C. Kesselman, (eds.). Computational Grids: The Future of High Performance Distributed Computing. Morgan Kaufman, San Mateo, CA., 1998.

....scalability at all. Other multi agent frameworks rely on another framework to solve the problem of scalability. For example, scalablity in the CoABS (DARPA Control of Agent Based Systems) approach [32] assumes adequate support from computational grids in providing a plug in backplane for agents [11]. In other multi agent frameworks, aspects of scalability are specifically addressed. In ZEUS [38] scalability is defined to be the growth rate of the maximum communication load (as a function of the number of agents) Their conclusions are that the maximum communication load grows at worst ....

I. Foster and C. Kesselman, (eds.). Computational Grids: The Future of High Performance Distributed Computing. Morgan Kaufman, San Mateo, CA., 1998.

....the prototype grid, in order to evaluate the utility of our technologies and of the grid concept. In this paper, we describe the status of the Globus project in each of these three areas, as of early 1998. This description updates the original Globus paper [13] and a subsequent project summary in [14] by providing a more complete and up to date description of the Globus toolkit and by reviewing early experiments with the Globus Ubiquitous Supercomputing Testbed (GUSTO) grid prototype, the largest computational grid constructed to date. The organization of this paper is as follows. In the next ....

I. Foster and C. Kesselman, editors. Computational Grids: The Future of High-Performance Distributed Computing. Morgan Kaufmann Publishers, 1998.

....the prototype grid, in order to evaluate the utility of our technologies and of the grid concept. In this paper, we describe the status of the Globus project in each of these three areas, as of early 1998. This description updates the original Globus paper [13] and a subsequent project summary in [14] by providing a more complete and up to date description of the Globus toolkit, and by reviewing early experiments with the Globus Ubiquitous Supercomputing Testbed (GUSTO) grid prototype, the largest computational grid constructed to date. The organization of this paper is as follows. In the next ....

I. Foster and C. Kesselman, editors. Computational Grids: The Future of High-Performance Distributed Computing. Morgan Kaufmann Publishers, 1998.

.... environment, called Delphi that leverages software from current projects at Illinois, Indiana, Wisconsin, and Argonne, including the Pablo [5] Paradyn [4] and Autopilot [6] performance analysis and measurement environments, the HPC and Polaris [1] Fortran compilers, and the Globus [2] metacomputing system. Delphi s goal is to support measurement, analysis, and performance prediction of multilingual, SPMD, data parallel, and object oriented applications executing on both homogeneous, parallel systems and distributed collections of wide area computing resources (i.e. ....

....CORBA ORB Paradyn Pablo Autopilot Virtue Nexus Globus Substrate Figure 1. Delphi Organizational Structure diction, and resource optimization for heterogeneous metacomputing. As Figure 1 suggests, Delphi integrates both parallel systems and wide area metacomputing or computational grids [2] within a single performance measurement and prediction framework. In this framework, annotating compilers emit detailed data on program transformations and annotate programs with calls to resource re ectors. Using this information, symbolic performance models can be used to rapidly assess the ....

[Article contains additional citation context not shown here]

I. Foster and C. Kesselman. Computational Grids: The Future of High-Performance Distributed Computing. Morgan-Kaufmann, 1998.

....computing and develops a security policy and a corresponding security architecture. An implementation of the architecture within the Globus metacomputing toolkit is discussed. 1 Introduction Large scale distributed computing environments, or computational grids as they are sometimes termed [4], couple computers, storage systems, and other devices to enable advanced applications such as distributed supercomputing, teleimmersion, computer enhanced instruments, and distributed data mining [2] Grid applications are distinguished from traditional client server applications by their ....

....deployment that this architecture is workable. 2 The Grid Security Problem We introduce the grid security problem with an example illustrated in Figure 1. This example, although somewhat contrived, captures important elements of real applications, such as those discussed in Chapters 2 5 of [4]. Data Data Data Data Data 1. Request data analysis 2. Contact resource broker 3. Initiate task farm 4. Access parameter values A. Physicist Kerberos physicist ssh ap SSL guest29 plaintext aphysicist plaintext ap6 plaintext bcollab Site A Site B Site C Site D Site E Site F Site G ....

I. Foster and C. Kesselman, editors. Computational Grids: The Future of High Performance Distributed Computing. Morgan Kaufmann, 1998.

....used to construct a second generation wide area MPI. 1 Introduction Wide area supercomputing or metacomputing environments couple geographically distributed resources to provide broader access to supercomputing capabilities or to enable qualitatively new classes of high performance applications [2, 11, 12, 16]. These environments combine aspects of traditional distributed and parallel computing systems. Metacomputing systems, like distributed systems, must deal with heterogeneity and dynamic behaviors; as in parallel computing, performance requirements often demand careful structuring of computation ....

I. Foster and C. Kesselman, editors. Computational Grids: The Future of High-Performance Distributed Computing. Morgan Kaufmann Publishers, 1998.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC