| P. Dinda. Online Prediction of the Running Time of Tasks. Cluster Computing, 5(3):225--236, July 2002. |
....from past measurements is present in many systems. Usually it is present implicitly in a feedback loop: measurements of load or performance are used as control signals to adjust system parameters [34] A few systems use explicit prediction of load: for example, Dinda s Running Time Advisor [10]. Explicit prediction of resource demand, however, is comparatively rare. Most systems assume that resource demand is constant, specified by the application, derived from a static probability distribution [23, 19] or obtained from compile time analysis [33] We know of two systems that ....
P. A. Dinda. Online Prediction of the Running Time of Tasks. In Proc. 10th IEEE International Symposium on High Performance Distributed Computing (HPDC '01), pages 383--394, San Francisco, CA, Aug. 2001.
....prediction is implicit in these systems: offered load or other system indicators are used as control signals that drive load balancing [58] scheduling [27] process migration [28, 45] or resource allocation [89] A few systems use explicit resource prediction. The Running Time Advisor [25] predicts the running time of a job on any host in a distributed system, given its nominal execution time: i.e. it predicts resource supply and performance, but the application must specify the resource demand. The RTA is based on Dinda et al. s work on host load prediction [23] using a variety ....
Peter A. Dinda. Online prediction of the running time of tasks. In Proceedings of the 10th IEEE International Symposium on High Performance Distributed Computing (HPDC '01), pages 383--394, San Francisco, CA, August 2001. IEEE Computer Society.
....as well as the proof (under some assumptions) of the validity of the reconstruction algorithm. Section 5 concludes this paper and presents future work. 2 State of the art Most of the previous work focused on collecting information about each separate host, or host to host communication [WSH99, Din02] and little attention has been given to gathering informations about the global network topology, and its impact on parallel data transfers. Most often, Grid systems rely on manual configuration [CD98] which is very error prone. Some tools [dBKB02, BCW] rely on the traceroute tool to get ....
Peter Dinda. Online prediction of the running time of tasks. Cluster Computing, 5(3), 2002.
....availability, and policy. Here, information sources are computers, and information can include both relatively static information such as system configuration (architecture, OS version, access policy) and more dynamic information such as instantaneous load and predictions of future availability [40, 10]. A replica selection service within a data grid responds to requests for the best copy of files that are replicated on multiple storage systems. Here, information sources can once again include system configuration, instantaneous performance, and predictions, but for storage systems and ....
P. Dinda. Online prediction of the running time of tasks. In Proc. 10th IEEE Symp. on High Performance Distributed Computing, (to appear) 2001.
No context found.
DINDA, P. A. Online prediction of the running time of tasks. Cluster Computing 5, 3 (2002.
No context found.
DINDA, P. A. Online prediction of the running time of tasks. Cluster Computing (2002.
No context found.
DINDA, P. A. Online prediction of the running time of tasks. Cluster Computing (2002.
No context found.
DINDA, P. A. Online prediction of the running time of tasks. Cluster Computing 5, 3 (2002.
No context found.
DINDA, P. A. Online prediction of the running time of tasks. Cluster Computing 5, 3 (2002).
....such as Globus [10] and Legion [13] simplify using remote resources within a computational grid. To help users find resources, these systems typically provide some form of a grid information service (GIS) such as Globus MDS [4] Resource monitoring tools such as NWS [23] Remos [5] or RPS [7] can be used to gauge the dynamic availability of found resources. Designing and evaluating such grid middleware demands realistic workloads. For example, we are in the process of designing and building a grid information service based Effort sponsored by the National Science Foundation under ....
P. A. Dinda. Online prediction of the running time of tasks. Cluster Computing, 5(3), 2002.
....[24] and network management [9] Our own focus is on providing applicationlevel performance queries to adaptive applications. For example, an application can ask the Running Time Advisor (RTA) system to predict, as a confidence interval, the running time of a given size task on a particular host [14]. We are trying to develop an analogous Message Transfer Time Advisor (MTTA) that, given two endpoints on an IP network, a message size, and a transport protocol, will re Effort sponsored by the National Science Foundation under Grants ANI 0093221, ACI 0112891, and EIA 0130869. The NLANR PMA ....
DINDA, P. A. Online prediction of the running time of tasks. Cluster Computing (2002.
....task s CPU demands, predictions of the load on a host, and a confidence level, the RTA predicts, as a confidence interval, the running time of the task on the host. The details of load measurement, prediction, and how the RTA computes its predictions of running time have been thoroughly documented [4, 6, 7, 5]. The Network Weather Service also provides load prediction [18] The RTSA is similar in spirit to the focused addressing algorithm described by Ramamritham, et al. [15] but it is based on sophisticated prediction techniques, makes no assumptions about host cooperation, and is designed to run on ....
....are submitted. The measurement hosts are the hosts from which the RTSA must choose. Each measurement host runs the following components: the load playback tool, a load sensor, one or more prediction systems, and a cycle server. The configuration of the prediction systems is discussed elsewhere [7, 5]. The cycle server runs tasks it takes requests to compute (using a busy loop) for some number of CPU seconds and then returns the wallclock time that the task took to complete. To evaluate the RTSA given a particular scenario, we started up the infrastructure with a particular scenario, ....
P. A. Dinda. Online prediction of the running time of tasks. Cluster Computing, 2002.
No context found.
P. Dinda. Online Prediction of the Running Time of Tasks. Cluster Computing, 5(3):225--236, July 2002.
No context found.
P. Dinda. Online Prediction of the Running Time of Tasks. Cluster Computing, 5(3):225--236, July 2002.
No context found.
P. Dinda. Online Prediction of the Running Time of Tasks. Cluster Computing, 5(3):225--236, July 2002.
No context found.
P. A. Dinda. Online Prediction of the Running Time of Tasks. In Proc. 10th IEEE International Symposium on High Performance Distributed Computing (HPDC '01), pages 383--394, San Francisco, CA, Aug. 2001.
No context found.
Dinda, P.A., Online Prediction of the Running Time of Tasks, Cluster Computing, 5 (2002).
No context found.
DINDA, P. A. Online prediction of the running time of tasks. Cluster Computing 5, 3 (2002.
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