Co-Scheduling Computational and Networking Resources in E-Science Optical Grids

Abouelela, Mohamed; El-Darieby, Mohamed

doi:10.1109/glocom.2010.5683835

Cited by 8 publications

(6 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Constraint (6) states that demand r traverses node i, if edge e : i → j is an edge traversed by r. Constraints (7) and (8) together ensure that the router at node i is active if i is selected as the destination node for at least one demand. Constraints (9), (10) and (11) together ensure that the optical switch at node i is active if there is at least one lightpath traversing node i, or i is selected as the destination node for at least one demand.…”

Section: Minimizementioning

confidence: 99%

“…In [5], the authors present a tabu search based approach for the joint optimization problem, where tasks are modeled as directed acyclic graphs (DAG). In [6], a genetic algorithm has been introduced for the task scheduling problem with the objective of minimizing both data processing time and data transfer time. The problems of resource reservation for sliding grid demands in optical grids are addressed in [7], [8].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Energy optimization in optical grids through anycasting

Chen

Jaekel

2013

2013 IEEE International Conference on Communications (ICC)

View full text Add to dashboard Cite

Optical grids are emerging as a natural, costeffective platform to meet the needs for powerful computing, large storage capacity and high-speed data transmission capabilities in a number of important application areas. In spite of the lower power cost per bit of optical networks, it is expected that one of the most challenging issues in the next decade will be reducing the power requirement for such core networks. Much of the recent work on optical grid network design has focused on optimizing the use of traditional computing and network resources in an integrated manner. In grid systems, it is typically possible to select one out of a number of possible destinations to execute a specific job. This is known as anycasting, and in this paper we propose a new approach for energy minimization in optical grids that exploits the inherent flexibility of anycasting. We present a comprehensive integer linear program (ILP) formulation that selects the destination node and performs routing and wavelength assignment (RWA) to minimize the overall energy consumption of a set of static lightpath demands. We also present a 2-stage ILP that can quickly generate solutions for large networks. Simulation results indicate that significant energy savings can be achieved by the proposed approach, not only compared to traditional RWA techniques but also over energy-aware unicast methods.

show abstract

Section: Minimizementioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Energy optimization in optical grids through anycasting

Chen

Jaekel

2013

2013 IEEE International Conference on Communications (ICC)

View full text Add to dashboard Cite

show abstract

“…Resource allocation in high performance grid computing is an area of ongoing research and development. Many researches were conducted illustrating the joint allocation approach and showing the advantage of it over the separated one [2][3][4][5][6]. Most of these efforts assume a centralized resource manager that has a complete vision of network topology as well as networking and computing resources status.…”

Section: Related Workmentioning

confidence: 99%

Multidomain Hierarchical Resource Allocation for Grid Applications

Abouelela

El-Darieby

2012

Journal of Electrical and Computer Engineering

Self Cite

View full text Add to dashboard Cite

Geographically distributed applications in grid computing environments are becoming more and more resource intensive. Many applications require the collaboration between different domains, may be independently administrated domains, to exchange data and share computing and storage resources. This collaboration should be done in a way that maintains the privacy of each participant domain. This calls for new architectures and approaches to deal with such multidomain environments. We propose a hierarchical-based architecture as well as multidomain hierarchical resource allocation approach. The resource allocation is performed in a distributed way among different domains such that each participant domain keeps its internal topology and private data hidden while sharing abstracted information with other domains. Both computing and networking resources are jointly scheduled while optimizing the application completion time taking into account data transfer delays. Simulation results show the scalability and feasibility of the proposed approach.

show abstract

“…Divisible Load Theory (DLT) has been successfully applied to parallel and distributed systems, as well as to grid computing environment [4], [5], [6]. Genetic Algorithms (GA) based approaches were also proposed to schedule Divisible Loads [7], [8]. Integer Linear Programming has also been introduced to model such problems [6].…”

Section: Related Workmentioning

confidence: 99%

“…None of the above contributions considers networking resources availability and connectivity while scheduling both computational and networking resources using DLT. In [7], a GA based approach that co-schedules computational and networking resources, while considering network resources availability and connectivity was introduced. The main drawback of this approach is the long GA execution time.…”

Section: Related Workmentioning

confidence: 99%