When we first decided that we wanted to buy a new cluster we got quotes from several different vendors. These quotes varied up to 50% for the same configuration based upon the addon features that each company provided. However, even the lowest figure was 30% more than what we calculated if we built the cluster ourselves. Once we decided that the 30% cost differential was significant enough to m...

Energy efficiency in databases is an emerging topic. Our research prototype WattDB dynamically adjusts the number of active servers in a cluster to the current workload to achieve energy proportionality. In this demo, we give insights in the partitioning process and WattDB’s reaction to workload changes by live-presenting a monitoring GUI. The whole process and the resulting configuration are v...

The behavior of a rainwater harvesting system depends on some variables that cannot be controlled, such as rainfall, building roof size and water consumption. selection regarding tank-size will affect the performance cost–benefit ratio. criterion employed for this is based need volume-storage typically, yield large-sized tanks, especially when amount higher during rainy seasons. This article pr...

In this paper we propose an accurate model of the population of mobile users in a cell cluster of an ATM wireless access network. Taking into account the statistical characteristics of user mobility and the network configuration (topology, the number of cells and radio channels in each cell), this model is able to capture the cross-correlation of the number of mobile users present in different ...

In order to overcome the energy loss involved by communications in wireless sensor networks (WSN), the use of clustering has proven to be effective. In this paper, we proposed a dynamic centralized genetic algorithm (GA)-based clustering approach to optimize the clustering configuration (cluster heads and cluster members) to limit node energy consumption. The obtained simulation results show th...

In very large clusters it is not possible to get Ethernet-switches that are large enough to support the whole cluster, thus a configuration with multiple switches are needed. This work seeks to evaluate the interconnection strategies for a new 300+ CPU cluster at the University of Southern Denmark. The focal point is a very inexpensive switch from D-Link which unfortunately offers only 24 Gb po...

UNLABELLED We have developed two programs that speed up common bioinformatic applications by spreading them across a UNIX cluster.(1) BLAST.pm, a new module for the 'MOLLUSC' package. (2) WRAPID, a simple tool for parallelizing large numbers of small instances of programs such as BLAST, FASTA and CLUSTALW. AVAILABILITY The packages were developed in Perl on a 20-node Linux cluster and are pro...