Elisa Schaeffer

Adaptive energy efficient scheduling in Peer-to-Peer desktop grids

Abstract We address non-preemptive scheduling problems on heterogeneous P2P grids, where resources are changing over time, and scheduling decisions are free from information of application characteristics. We consider a scheduling with task replications to overcome possible bad resource allocation in presence of uncertainty, and ensure good performance. We analyze the energy consumption of job allocation strategies exploring the replication thresholds, and dynamic component deactivation. The main idea of our approach is to set replication thresholds, and dynamically adapt them to cope with different objective preferences, workloads, and Grid properties. We compare three groups of strategies: knowledge-free, speed-aware, and power-aware. In order to provide good performance and minimize energy consumption, first, we perform a joint analysis of two metrics considering their degradation in performance. Then, we provide two-objective optimization analysis that is not restricted to find a unique solution, but the Pareto optimal set. Based on these results, we use a Set Coverage metric for assessing the performance of our strategies and compare twenty algorithms in terms of Pareto dominance. A case study is given, and corresponding results indicate that two replicas for knowledge-free algorithms, and one replica for speed-aware algorithms provide the best energy and performance trade-offs in the scheduling. They perform well in different scenarios with a variety of workloads and grid configurations.

Energy efficiency of knowledge-free scheduling in Peer-to-Peer Desktop Grids

We address knowledge-free Bag-of-Tasks non-preemptive scheduling problem on heterogeneous grids, where scheduling decisions are free from information of resources and application characteristics. We consider a scheduling with task replications to overcome possible random bad resource allocation and ensure good performance. We analyze energy consumption of job allocation strategies based on variations of the replication threshold. In order to provide QoS and minimize energy consumption, we perform a joint analysis of two metrics. A case study is given and corresponding results indicate that proposed strategies reduce energy consumption without significant degradation in performance.

Local Survival Rule for Steer an Adaptive Ant-Colony Algorithm in Complex Systems

The most prevalent P2P application today is file sharing, both among scientific users and the general public. A fundamental process in file sharing systems is the search mechanism. The unstructured nature of real-world large-scale complex systems poses a challenge to the search methods, becasuse global routing and directory services are impractical to implement. In this paper, a new ant-colony algorithm, Adaptive Neighboring-Ant Search (AdaNAS), for the semantic query routing problem (SQRP) in a P2P network is presented. The proposed algorithm incorporates an adaptive control parameter tuning technique for runtime estimation of the time-to-live (TTL) of the ants. AdaNAS uses three strategies that take advantage of the local environment: learning, characterization, and exploration. Two classical learning rules are used to gain experience on past performance using three new learning functions based on the distance traveled and the resources found by the ants. The experimental results show that the AdaNAS algorithm outperforms the NAS algorithm where the TTL value is not tuned at runtime.

Improving Distributed Resource Search through a Statistical Methodology of Topological Feature Selection

The Internet is considered a complex network for its size, interconnectivity and rules that govern are dynamic, because of constantly evolve. For this reason the search of distributed resources shared by users and online communities is a complex task that needs efficient search method. The goal of this work is to improve the performance of distributed search of information, through analysis of the topological features. In this paper we described a statistical methodology to select a set of topologic metrics that allow to locally distinguish the type of complex network. In this way we use the metrics to guide the search towards nodes with better connectivity. In addition we present an algorithm for distributed search of information, enriched with the selected topological metric. The results show that including the topological metric in the Neighboring-Ant Search algorithm improves its performance 50% in terms of the number of hops needed to locate a set of resources. The methodology described provides a better understanding of why the features were selected and aids to explain how this metric impacts in the search process.