GridBench: A tool for the interactive performance exploration of Grid infrastructures

Tsouloupas, George; Dikaiakos, Marios D.

doi:10.1016/j.jpdc.2007.04.009

Cited by 21 publications

(9 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They test and measure performance of basic Grid functions, including file transfers, remote execution, and Grid Information Services response. GridBench [ 9] is a complex and interesting tool that provides a graphical interface to define, execute and administrate benchmarks. It takes into account interconnection performance and evaluates resource workload, and can be used to rank Grid resources.…”

Section: Related Workmentioning

confidence: 99%

“…Both benchmarks have been run on a CPU/core independently, and then the aggregated results are gathered to represent the performance of the whole parallel resources [ 9]. For each resource, we present the evaluation of the single CPU/core and the parallel resources.…”

Section: Fig 1 Comparison Between Resources According To Flopsmentioning

confidence: 99%

“…MM merges these lists with its own PO list and produce a Global List, ordered on the ranking values, that is passed to JS (7), which returns it back to RS (8). RS applies its scheduling policy to determine the resource to use, and calls the JS of the GREEN responsible of the PO owning the selected machine (GREEN PO 2 's instance in our case), by sending it the extended JSDL document along with the data indentifying the selected resource (9). This JS translates the information regarding the job execution of the original JSDL document in the format proper of the specific PO middleware, stating the resource on which the computation takes place (producing a JDD document for GT4 resources or a JDL document for the gLite ones), and finally, activates the Execution Environment in charge of executing the job represented in the translated document (10).…”

Section: Components Descriptionmentioning

confidence: 99%

“…The responsibility for performing this crucial activity is commonly left to the matchmaking component, whose main task is to grant the discovery of available resources and services on the base of the specific properties defined by users and expressed through pertinent requests [ 8]. To this end, the matchmaker may greatly benefit from a performance characterization of resources based on the employment of benchmarks [ 9].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Performance-Based Methodology to Improve Grid Exploitation

Clematis

Corana

D’Agostino

et al. 2009

Integrated Management of Systems, Services, Processes and People in IT

View full text Add to dashboard Cite

Abstract. Due to their complexity, the exploitation of Grid environments is not a trivial activity for many users, and a key factor is to enable a simplified and transparent orchestration of resources and jobs. Particularly critical is the deployment of matching procedures capable to effectively meet user's requirements with resources offer. We introduce GREEN a management tool primarily devoted to the matchmaking process, based on a performance characterization of both resources and job requirements. Leveraging on a two-level benchmarking methodology, GREEN allows users to express performance preference through an appropriate extension to Grid submission and description languages such as JSDL and Glue. Operating at intermediate level between applications and Grid middleware, GREEN reduces the gap between users' needs and available resources thus enabling a seamless exploitation of the Grid.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Fig 1 Comparison Between Resources According To Flopsmentioning

confidence: 99%

Section: Components Descriptionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Performance-Based Methodology to Improve Grid Exploitation

Clematis

Corana

D’Agostino

et al. 2009

Integrated Management of Systems, Services, Processes and People in IT

View full text Add to dashboard Cite

show abstract

“…In particular, one can utilize tools such as GridBench [26,28], the Grid Assessment Probes [4] and DiPerF [7], in order to determine in real time the value of certain low level and application-level failure semantics that can not be furnished by the meta-information sources. For example, the GridBench tool developed by our group provides a corpus of over 20 benchmarks that can be used to evaluate and rank the performance of Grid sites and individual Grid nodes.…”

Section: Monitoring Failures In a Grid Environmentmentioning

confidence: 99%

Metadata Ranking and Pruning for Failure Detection in Grids

Zeinalipour-Yazti

Papadakis

Georgiou

et al. 2008

Parallel Process. Lett.

Self Cite

View full text Add to dashboard Cite

The objective of Grid computing is to make processing power as accessible and easy to use as electricity and water. The last decade has seen an unprecedented growth in Grid infrastructures which nowadays enables large-scale deployment of applications in the scientific computation domain. One of the main challenges in realizing the full potential of Grids is making these systems dependable.In this paper we present FailRank, a novel framework for integrating and ranking information sources that characterize failures in a grid system. After the failing sites have been ranked, these can be eliminated from the job scheduling resource pool yielding in that way a more predictable, dependable and adaptive infrastructure. We also present the tools we developed towards evaluating the FailRank framework. In particular, we present the FailBase Repository which is a 38GB corpus of state information that characterizes the EGEE Grid for one month in 2007. Such a corpus paves the way for the community to systematically uncover new, previously unknown patterns and rules between the multitudes of parameters that can contribute to failures in a Grid environment. Additionally, we present an experimental evaluation study of the FailRank system over 30 days which shows that our framework identifies failures in 93% of the cases and can achieve this by only fetching 65% of the available information sources. We believe that our work constitutes another important step towards realizing adaptive Grid computing systems.

show abstract

Finding order in chaos: a behavior model of the whole grid

Montes

Sánchez

Valdés

et al. 2010

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARYOver the last decade, grid computing has paved the way for a new level of large scale distributed systems where thousands or even millions of heterogeneous, geographically distributed resources can join to provide vast cuantities of integrated computational resources. This powerful capabilities can be used to address computationally challenging scientific problems that could not be faced before, but also provide a new environment where more commercially-oriented ideas could be deployed, such as utility computing or the more recent cloud computing. The grid computing technology developed in the last fifteen years is the natural solution for the developing of these initiatives.But this new step in distributed computing presents also a completely new level of complexity. Management mechanisms play a key role in making grid computing possible, and the correct analysis and understanding of the grid behavior is needed to assure its optimal performance. Usual distributed computing management mechanisms analyze * Correspondence to: jmontes@cesvima.upm.es FINDING ORDER IN CHAOS: A BEHAVIOR MODEL OF THE WHOLE GRID1 each resource as part of the system and improve global performance by adjusting specific parameters of all of them. Nevertheless, when trying to adapt the same procedures to grid computing the vast heterogeneity and complexity of the system complicates this task.On the other hand, grid heterogeneity and complexity could only be a matter of perspective. Maybe it is possible to analyze and understand the grid behavior as a single system, instead of a set of thousands of them. This abstraction could provide a deeper understanding of the system, describing large scale patterns and global events that could not be detected (or it would be very difficult to) analyzing each resource in a separate way. The subsequent synergy could play the key role in taking grid performance to its peak.In this paper a specific set of techniques is presented and described in order to create a global behavior model of the grid, analyzing it as a single entity. Several real and simulated study cases are also presented, in order to provide a proper validation and illustrate the benefits of this approach.

show abstract

GridBench: A tool for the interactive performance exploration of Grid infrastructures

Cited by 21 publications

References 7 publications

A Performance-Based Methodology to Improve Grid Exploitation

A Performance-Based Methodology to Improve Grid Exploitation

Metadata Ranking and Pruning for Failure Detection in Grids

Finding order in chaos: a behavior model of the whole grid

Contact Info

Product

Resources

About