Multidomain Hierarchical Resource Allocation for Grid Applications

Abstract

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.

Specification and Design of Trust-Based Open Self-Organising Systems

“…Given an adequate hierarchical structure, this results in shorter scheduling times for the overall system in comparison to a centralised approach. While hierarchical problem decomposition has been proposed as a generic approach to deal with large-scale systems prohibiting a centralised solution [18,19], our self-organising hierarchy of AVPPs is a means to autonomously decompose the scheduling problem at runtime. This allows the AVPPs to adapt their organisational structure in response to changes in the environment or the agents' internal state, and thereby to maintain an appropriate compromise between solution quality and runtime performance.…”

An Approach to Robust Resource Allocation in Large-Scale Systems of Systems

“…Regarding scalability, hierarchical problem decomposition has been proposed as a generic approach to deal with large-scale systems prohibiting a centralized solution [10], [11]. As we showed in [12], agents can establish and maintain appropriate hierarchical system structures using self-organization.…”

Synthesis and Abstraction of Constraint Models for Hierarchical Resource Allocation Problems

“…Resource allocation problems present themselves in a variety of domains (Chevaleyre et al, 2006), including autonomous power management (Anders et al, 2013b) or grid computing (Abouelela and El-Darieby, 2012). Constraint programming provides a wide range of algorithms and techniques targeted at solving and optimising problems including resource allocation that are readily available in state-of-the-art software.…”

“…This proves particularly useful when the characteristics of heterogeneous agents need to be modelled. If, however, the size of the system prohibits a centralised solution, due either to the communication overhead required in collecting all necessary information or due to the complexity of a centralised solution model, hierarchical decomposition provides a generic approach to deal with these issues (see, e.g., Abouelela and El-Darieby, 2012;Boudjadar et al, 2013). In such cases, distributed, cooperative, agent-based systems can be employed in which the agents form organisations.…”