Cluster Computing and Genetic Algorithms With Cluster Knoppix

Cluster Computing and Genetic Algorithms With Cluster. Knoppix David Tabachnick

What is a Cluster? A Cluster is… A group of connected computers, acting as one to calculate computationally difficult problems. A Cluster is… Made up of a server, nodes, and networking hub(s).

Example of a Fully Developed Cluster

Basics of Clustering § Server is the head of the cluster. § Nodes are all connected via hub. § Server is started first. The nodes are then powered on, and they look to the server for instructions. § The cluster is both scalable and highly available. § The cluster can be taken down and restarted quickly and easily.

Methods to Clustering § Load-balancing is done on the server by a software package, Open. Mosix. § This results in breaking programs into processes that can be sent to the nodes for processing. § No special programming methods or languages must be used.

Features of this Cluster § Live Install with configuration scripts § Remote access for authorized users § High scalability § Extremely secure § Local storage on the server and external storage for redundant data backup § Will run LISP

Genetic Algorithms on a Cluster § Server performs the string manipulation and sends the matrix calculations to the nodes. § Each node is given one fitness function to calculate. § Upon completion, a node will write back the resultant function and remain idle. § ≈80% of the calculations take place on the nodes.

Genetic Algorithm Clustering

Conclusion § This cluster… § Achieved relatively high performance when compared to other clusters of similar size and scope. § Was designed with computing genetic algorithms in mind. § Was built to be scalable, flexible, and easily accessible.