| Montresor, A., Meling, H., Babaoglu, O.: Messor: Load-Balancing through a Swarm of Autonomous Agents. In: Proceedings of the International Workshop on Agents and Peer-toPeer Computing in conjunction with AAMAS 2002, Bologna, Italy (2002) |
....specified using XML. After having developed a prototype of Anthill, we are now in the process of testing the viability of our ideas by developing CAS based P2P applications, including a load balancing algorithm for grid computing based on the dispersing ant described in the previous section [6] and a document sharing application based on keyword pheromone trails left by exploring ants [1] Preliminary simulation results appear to confirm our intuition that applying CAS based techniques to solving challenging problems in distributed systems is a promising approach. The performance of our ....
A. Montresor, H. Meling, and O. Babaoglu. Messor: Load-Balancing through a Swarm of Autonomous Agents. In Proc. of the 1st Workshop on Agent and Peerto -Peer Systems, Bologna, Italy, July 2002.
....10 100 1000 10000 1 100 (a) b) c) d) e) f) Fig. 5. Load distribution after 0, 5, 10, 15, 20, 50 iterations. 4 Performance Evaluation In this section, we show some preliminary results about Messor obtained through the Anthill simulator. More details can be found in a companion paper [16]. Figure 5 shows the process of load balancing performed by Messor over time. These results have been obtained in a network of 100 idle nests, initially connected to form a ring (for visualization reasons) At the beginning, 10,000 jobs are generated in a single node. The di#erent histograms ....
A. Montresor, H. Meling, and A. Montresor. Messor: Load-Balancing through a Swarm of Autonomous Agents. Technical Report UBLCS-02-08, Dept. of Computer Science, University of Bologna, May 2002.
....of computers lacks xed structures as its nodes come and go freely. Therefore the master slave structure, where a centralized master disperses computing tasks to slaves, is not applicable to load balancing on grid. Complex adaptive systems (CAS) are considered to be suitable to the situation [4, 5, 1]. CAS are usually used to describe the dynamic and evolutionary behavior in biological and social systems. In CAS, single agents show simple strategies in isolation. However, as a large number of agents interact with each other, the system exhibits richer and more complex behavior than those ....
....object, it picks the object up and continues to wander. While each agent follows the rules, Resnick shows that an ant colony can collect objects into piles. The result of ant s strategies in [8] is collecting objects. Results of the inverse strategies is dispersing objects, which is considered in [5, 1]. In [5] a Peer to Peer (P2P) paradigm is designed to execute load balancing: ants which carry objects leave their objects behind only after they have wandered about randomly for a while without encountering object concentrations. As a result, objects are dispersed uniformly rather than collected ....
[Article contains additional citation context not shown here]
A. Montresor and H. Meling. Messor:load-balancing through a swarm of autonomous agents. Technical Report , Dept. of Computer Science, University of Bologna, UBLCS-02-08, May 2002.
....10 100 1000 10000 1 100 (a) b) c) d) e) f) Fig. 5. Load distribution after 0, 5, 10, 15, 20, 50 iterations. 4 Performance Evaluation In this section, we present preliminary results for Messor, obtained through the Anthill simulator. Further details can be found in a companion paper [17]. Figure 5 illustrate how the load balancing process performed by Messor evolves over time. The results were obtained in a network of 100 idle nests, initially connected to form a ring (for visualization reasons) Initially, 10,000 jobs are generated in a single node. The di#erent histograms ....
A. Montresor, H. Meling, and O. Babaoglu. Messor: Load-Balancing through a Swarm of Autonomous Agents. Technical Report UBLCS-02-08, Dept. of Computer Science, University of Bologna, May 2002. In preparation.
....(in SearchMax) or underloaded (in SearchMin) In this way, ants move faster towards those regions of the network in which they are more interested. Figure 1 shows some preliminary results obtained by Messor. More details about the algorithm and its performance can be found in a companion paper [6]. These results were obtained in a network of 100 idle nodes, by generating 100,000 tasks from a single node. At each iteration, all ants perform a single step by executing their algorithm and moving to the next node. As shown in the figure, after only 40 iterations, the workload is evenly ....
A. Montresor, H. Meling, and A. Montresor. Messor: Load-Balancing through a Swarm of Autonomous Agents. Technical Report UBLCS-02-08, Dept. of Computer Science, University of Bologna, May 2002. In preparation.
No context found.
Montresor, A., Meling, H., Babaoglu, O.: Messor: Load-Balancing through a Swarm of Autonomous Agents. In: Proceedings of the International Workshop on Agents and Peer-toPeer Computing in conjunction with AAMAS 2002, Bologna, Italy (2002)
No context found.
A. Montresor, H. Meling, and O. Babaoglu. Messor: LoadBalancing through a Swarm of Autonomous Agents. In Proceedings of the 1st Workshop on Agent and Peer-to-Peer Systems, July 2002.
No context found.
A. Montresor, H. Meling, and O. Babaoglu. Messor: LoadBalancing through a Swarm of Autonomous Agents. In Proceedings of the 1st Workshop on Agent and Peer-to-Peer Systems, July 2002.
No context found.
A. Montresor, H. Meling, and O. Babaoglu. Messor: LoadBalancing through a Swarm of Autonomous Agents. In Proceedings of the 1st Workshop on Agent and Peer-to-Peer Systems, July 2002.
No context found.
A. Montresor, H. Meling, and O. Babaoglu, "Messor: Load-Balancing through a Swarm of Autonomous Agents," International Workshop on Agents and Peer-to-Peer Computing, 2002.
No context found.
Montresor A, Meling H et al, "Messor: Load-Balancing through a Swarm of Autonomous Agents", Proceedings 1 st International Workshop on Agents and P2P Computing, Bologna, Italy, July 2002
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC