A WORKLOAD CHARACTERIZATION BY CLUSTERING TECHNIQUE

An improvement of a method for the characterization of computer worklo ad by means of arrival patterns is presented. A numerical fitting tech nique provides a suitable representation of the arrival rate function of jobs over one day period. In order to classify such arrival pattern s, we suggest the application of the Mac Queen algorithm with coarseni ng and refining parameters [2], which doesn't need the number of clust ers to be fixed a priori, as in the original approach [4]. Solutions f or some numerical problems related to the approximation of the arrival rate function are also provided. The proposed technique has been impl emented and experimental results concerning the workload characterizat ion of an educational system are given.