The Experts below are selected from a list of 67728 Experts worldwide ranked by ideXlab platform

Nagarajan Kandasamy - One of the best experts on this subject based on the ideXlab platform.

  • approximation modeling for the online performance management of Distributed Computing Systems
    Systems Man and Cybernetics, 2008
    Co-Authors: Dara Kusic, Nagarajan Kandasamy, Guofei Jiang
    Abstract:

    A promising method of automating management tasks in Computing Systems is to formulate them as control or optimization problems in terms of performance metrics. For an online optimization scheme to be of practical value in a Distributed setting, however, it must successfully tackle the curses of dimensionality and modeling. This paper develops a hierarchical control framework to solve performance management problems in Distributed Computing Systems operating in a data center. Concepts from approximation theory are used to reduce the computational burden of controlling such large-scale Systems. The relevant approximations are made in the construction of the dynamical models to predict system behavior and in the solution of the associated control equations. Using a dynamic resource-provisioning problem as a case study, we show that a Computing system managed by the proposed control framework with approximation models realizes profit gains that are, in the best case, within 1% of a controller using an explicit model of the system.

  • approximation modeling for the online performance management of Distributed Computing Systems
    International Conference on Autonomic Computing, 2007
    Co-Authors: Dara Kusic, Nagarajan Kandasamy, Guofei Jiang
    Abstract:

    This paper develops a hierarchical control framework to solve performance management problems in Distributed Computing Systems. To reduce the control overhead, concepts from approximation theory are used in the construction of the dynamical models that predict system behavior, and in the solution of the associated control equations themselves. Using a dynamic resource provisioning problem as a case study, we show that a Computing system managed by the proposed control framework using approximation models realizes profit gains that are, in the best case, within 1% of a controller using an exact parametric model of the system.

  • a hierarchical optimization framework for autonomic performance management of Distributed Computing Systems
    International Conference on Distributed Computing Systems, 2006
    Co-Authors: Nagarajan Kandasamy, Sherif Abdelwahed, Mohit Khandekar
    Abstract:

    This paper develops a scalable online optimization framework for the autonomic performance management of Distributed Computing Systems operating in a dynamic environment to satisfy desired quality-ofservice objectives. To efficiently solve the performance management problems of interest in a Distributed setting, we develop a hierarchical structure where a highlevel limited-lookahead controller manages interactions between lower-level controllers using forecast operating and environment parameters. We develop the overall control structure, and as a case study, show how to efficiently manage the power consumed by a computer cluster. Using workload traces from the Soccer World Cup 98 web site, we show via simulations that the proposed method is scalable, has low run-time overhead, and adapts quickly to time-varying workload patterns.

  • ICDCS - A Hierarchical Optimization Framework for Autonomic Performance Management of Distributed Computing Systems
    26th IEEE International Conference on Distributed Computing Systems (ICDCS'06), 1
    Co-Authors: Nagarajan Kandasamy, Sherif Abdelwahed, Mohit Khandekar
    Abstract:

    This paper develops a scalable online optimization framework for the autonomic performance management of Distributed Computing Systems operating in a dynamic environment to satisfy desired quality-ofservice objectives. To efficiently solve the performance management problems of interest in a Distributed setting, we develop a hierarchical structure where a highlevel limited-lookahead controller manages interactions between lower-level controllers using forecast operating and environment parameters. We develop the overall control structure, and as a case study, show how to efficiently manage the power consumed by a computer cluster. Using workload traces from the Soccer World Cup 98 web site, we show via simulations that the proposed method is scalable, has low run-time overhead, and adapts quickly to time-varying workload patterns.

Guofei Jiang - One of the best experts on this subject based on the ideXlab platform.

  • approximation modeling for the online performance management of Distributed Computing Systems
    Systems Man and Cybernetics, 2008
    Co-Authors: Dara Kusic, Nagarajan Kandasamy, Guofei Jiang
    Abstract:

    A promising method of automating management tasks in Computing Systems is to formulate them as control or optimization problems in terms of performance metrics. For an online optimization scheme to be of practical value in a Distributed setting, however, it must successfully tackle the curses of dimensionality and modeling. This paper develops a hierarchical control framework to solve performance management problems in Distributed Computing Systems operating in a data center. Concepts from approximation theory are used to reduce the computational burden of controlling such large-scale Systems. The relevant approximations are made in the construction of the dynamical models to predict system behavior and in the solution of the associated control equations. Using a dynamic resource-provisioning problem as a case study, we show that a Computing system managed by the proposed control framework with approximation models realizes profit gains that are, in the best case, within 1% of a controller using an explicit model of the system.

  • approximation modeling for the online performance management of Distributed Computing Systems
    International Conference on Autonomic Computing, 2007
    Co-Authors: Dara Kusic, Nagarajan Kandasamy, Guofei Jiang
    Abstract:

    This paper develops a hierarchical control framework to solve performance management problems in Distributed Computing Systems. To reduce the control overhead, concepts from approximation theory are used in the construction of the dynamical models that predict system behavior, and in the solution of the associated control equations themselves. Using a dynamic resource provisioning problem as a case study, we show that a Computing system managed by the proposed control framework using approximation models realizes profit gains that are, in the best case, within 1% of a controller using an exact parametric model of the system.

Mohit Khandekar - One of the best experts on this subject based on the ideXlab platform.

  • a hierarchical optimization framework for autonomic performance management of Distributed Computing Systems
    International Conference on Distributed Computing Systems, 2006
    Co-Authors: Nagarajan Kandasamy, Sherif Abdelwahed, Mohit Khandekar
    Abstract:

    This paper develops a scalable online optimization framework for the autonomic performance management of Distributed Computing Systems operating in a dynamic environment to satisfy desired quality-ofservice objectives. To efficiently solve the performance management problems of interest in a Distributed setting, we develop a hierarchical structure where a highlevel limited-lookahead controller manages interactions between lower-level controllers using forecast operating and environment parameters. We develop the overall control structure, and as a case study, show how to efficiently manage the power consumed by a computer cluster. Using workload traces from the Soccer World Cup 98 web site, we show via simulations that the proposed method is scalable, has low run-time overhead, and adapts quickly to time-varying workload patterns.

  • ICDCS - A Hierarchical Optimization Framework for Autonomic Performance Management of Distributed Computing Systems
    26th IEEE International Conference on Distributed Computing Systems (ICDCS'06), 1
    Co-Authors: Nagarajan Kandasamy, Sherif Abdelwahed, Mohit Khandekar
    Abstract:

    This paper develops a scalable online optimization framework for the autonomic performance management of Distributed Computing Systems operating in a dynamic environment to satisfy desired quality-ofservice objectives. To efficiently solve the performance management problems of interest in a Distributed setting, we develop a hierarchical structure where a highlevel limited-lookahead controller manages interactions between lower-level controllers using forecast operating and environment parameters. We develop the overall control structure, and as a case study, show how to efficiently manage the power consumed by a computer cluster. Using workload traces from the Soccer World Cup 98 web site, we show via simulations that the proposed method is scalable, has low run-time overhead, and adapts quickly to time-varying workload patterns.

Dara Kusic - One of the best experts on this subject based on the ideXlab platform.

  • approximation modeling for the online performance management of Distributed Computing Systems
    Systems Man and Cybernetics, 2008
    Co-Authors: Dara Kusic, Nagarajan Kandasamy, Guofei Jiang
    Abstract:

    A promising method of automating management tasks in Computing Systems is to formulate them as control or optimization problems in terms of performance metrics. For an online optimization scheme to be of practical value in a Distributed setting, however, it must successfully tackle the curses of dimensionality and modeling. This paper develops a hierarchical control framework to solve performance management problems in Distributed Computing Systems operating in a data center. Concepts from approximation theory are used to reduce the computational burden of controlling such large-scale Systems. The relevant approximations are made in the construction of the dynamical models to predict system behavior and in the solution of the associated control equations. Using a dynamic resource-provisioning problem as a case study, we show that a Computing system managed by the proposed control framework with approximation models realizes profit gains that are, in the best case, within 1% of a controller using an explicit model of the system.

  • approximation modeling for the online performance management of Distributed Computing Systems
    International Conference on Autonomic Computing, 2007
    Co-Authors: Dara Kusic, Nagarajan Kandasamy, Guofei Jiang
    Abstract:

    This paper develops a hierarchical control framework to solve performance management problems in Distributed Computing Systems. To reduce the control overhead, concepts from approximation theory are used in the construction of the dynamical models that predict system behavior, and in the solution of the associated control equations themselves. Using a dynamic resource provisioning problem as a case study, we show that a Computing system managed by the proposed control framework using approximation models realizes profit gains that are, in the best case, within 1% of a controller using an exact parametric model of the system.

Sheng-de Wang - One of the best experts on this subject based on the ideXlab platform.

  • Heuristic task assignment for Distributed Computing Systems
    Information Sciences, 1992
    Co-Authors: Chiun-chieh Hsu, Sheng-de Wang
    Abstract:

    Abstract This paper addresses the problem of assigning a task with precedence constraint to a Distributed Computing system. The task turnaround time including communication overhead and idle time is adopted to measure the performance of task assignment. The task assignment in this paper requires one to determine not only the assignment of modules, but also the sequence of message transmission to balance processor loads and diminish communication overhead. The search for the optimal task assignment with precedence constraint is known to be NP-complete in the strong sense. A heuristic algorithm with polynomial time complexity is then proposed in order to effectively solve the task assignment problem. The experimental results reveal that the proposed approach is able to obtain a near-optimal or even the optimal task assignment.