ReGS: user-level reliability in a grid environment

Sanches, J Afonso; Vargas, Patrícia Kayser; Dutra, Inês; Costa, Vítor Santos; Geyer, Cláudio F. R.

doi:10.1109/ccgrid.2005.1558634

Cited by 4 publications

(5 citation statements)

References 7 publications

(12 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(i) Once a task or set of tasks is allocated to a Grid node, it will not be scheduled again or at least this will be transparent to the higher-level resource managers. Although premise (i) seems a bit restrictive, GRAND employs a mechanism similar to that used in ReGS [12] to allow resubmission of failed tasks. Each TM checks for the correct termination of each task.…”

Section: The Grand Modelmentioning

confidence: 99%

“…As a consequence of the network congestion, we had corrupted output files. This forced us to adopt a more cautious procedure to submit our tasks to Condor, which resulted in the ReGS application management system [11,12], whose main objective was to resubmit failed tasks. Network flow and submission were still controlled manually (with the aid of shell scripts) with batch submission of a few tasks at a time.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

GRAND: toward scalability in a Grid environment

Vargas

Dutra

Nascimento

et al. 2006

Concurrency and Computation

View full text Add to dashboard Cite

One of the challenges in Grid computing research is to provide a means to automatically submit, manage, and monitor applications whose main characteristic is to be composed of a large number of tasks. The large number of explicit tasks, generally placed on a centralized job queue, can cause several problems: (1) they can quickly exhaust the memory of the submission machine; (2) they can deteriorate the response time of the submission machine due to these demanding too many open ports to manage remote execution of each of the tasks; (3) they may cause network traffic congestion if all tasks try to transfer input and/or output files across the network at the same time; (4) they make it impossible for the user to follow execution progress without an automatic tool or interface; (5) they may depend on fault-tolerance mechanisms implemented at application level to ensure that all tasks terminate successfully. In this work we present and validate a novel architectural model, GRAND (Grid Robust ApplicatioN Deployment), whose main objective is to deal with the submission of a large numbers of tasks.

show abstract

Section: The Grand Modelmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

GRAND: toward scalability in a Grid environment

Vargas

Dutra

Nascimento

et al. 2006

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…Each node corresponds to an application task, and each edge represents a task dependence that indicates a precedence order. However, in GRAND, each node does not need to be explicitly defined, because the application can be described through an iterator command, provided by our description language, GRID-ADL (Grid Application Description Language) [18]. A GRID-ADL description is the input file to GRAND.…”

Section: The Grand Modelmentioning

confidence: 99%

“…The coordination of the tasks was done through an user-level tool to control and monitor that specific application execution, including automatic resubmission of failed tasks. A later work in the same direction showed that an user-level tool such as that could also be useful to indicate how to select grid nodes to run a set of tasks [18]. These works only approached the problem of managing and monitoring tasks belonging to a specific application to help interaction with the user.…”

Section: Introductionmentioning

confidence: 99%

“…GRAND takes as input a simple application description language, called GRID-ADL, that can be obtained directly from the user or can be translated from other languages, including graphical representations. For example, the ReGS [18] tool has a web form that allows users to set some parameters for a machine learning application. From this form, a GRID-ADL file is automatically generated.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Hierarchical submission in a Grid environment

Vargas

Dutra

Nascimento

et al. 2005

Proceedings of the 3rd International Workshop on Middleware for Grid Computing - MGC '05

View full text Add to dashboard Cite

One of the challenges in grid computing research is to provide means to automatically submit, manage, and monitor applications which spread a large number of tasks. The usual way of managing these tasks is to represent each one as an explicit node in a graph, and this is the approach taken by many grid systems up to date. This approach can quickly saturate the machine where the application is launched, as we increase the number of tasks. In this work we present and validate a novel architectural model, GRAND (Grid Robust ApplicatioN Deployment), whose main objective is to deal with the problem of memory and load saturation of the submission machine. GRAND is implemented at a middleware level, aiming at providing a distributed task submission through a hierarchical organization. This paper provides an overview of the GRAND submission model as well our implementation. Initial results show that our approach can be much more effective than other approaches in the literature.

show abstract