Long live the queen: studying aging in social insects

Aging is a fascinating, albeit controversial, chapter in biology. Few other subjects have elicited more than a century of ever-increasing scientific interest. In this review, we discuss studies on aging in social insects, a group of species that includes ants and termites, as well as certain bee and wasp species. One striking feature of social insects is the lifespan of queens (reproductive females), which can reach nearly 30 years in some ant species. This is over 100 times the average lifespan of solitary insects. Moreover, there is a tremendous variation in lifespan among castes, with queens living up to 500 times longer than males and 10 times longer than workers (non-reproductive individuals). This lifespan polymorphism has allowed researchers to test the evolutionary theory of aging and Y more recently Y to investigate the proximate causes of aging. The originality of these studies lies in their use of naturally evolved systems to address questions related to aging and lifespan determination that cannot be answered using the conventional model organisms.

L'émigration marchande savoyarde aux XVIIe et XVIIIe siècles : l'exemple de Nancy-sur-Cluses / Chantal et Gilbert Maistre

Collection : Mémoires et documents publiés par l'Académie salésienne ; 94

Improving distributed simulation in a workstation environment

Technological development brings more and more complex systems to the consumer markets. The time required for bringing a new product to market is crucial for the competitive edge of a company. Simulation is used as a tool to model these products and their operation before actual live systems are built. The complexity of these systems can easily require large amounts of memory and computing power. Distributed simulation can be used to meet these demands. Distributed simulation has its problems. Diworse, a distributed simulation environment, was used in this study to analyze the different factors that affect the time required for the simulation of a system. Examples of these factors are the simulation algorithm, communication protocols, partitioning of the problem, distributionof the problem, capabilities of the computing and communications equipment and the external load. Offices offer vast amounts of unused capabilities in the formof idle workstations. The use of this computing power for distributed simulation requires the simulation to adapt to a changing load situation. This requires all or part of the simulation work to be removed from a workstation when the owner wishes to use the workstation again. If load balancing is not performed, the simulation suffers from the workstation's reduced performance, which also hampers the owner's work. Operation of load balancing in Diworse is studied and it is shown to perform better than no load balancing, as well as which different approaches for load balancing are discussed.