Generating hierarchical overlays for exponential topologies

P. VERGHELET; E. MOCSKOS

Córdoba

Jornada; HPC-Day 2013; 2013

Sociedad Argentina de Investigación Operativa - HPCLatAm

Grid computing refers to the federation of geographically distributed and heterogeneous computer resources.These resources may belong to different administrative domains, but are shared among users.Every grid presents a key component responsible for obtaining, distributing, indexing and archiving information about the configuration and state of services and resources.In large scale Grids, the dynamics of the resource information cannot be captured using a static hierarchy and relying in manual configuration and administration. It is necessary to design new policies for discovery and propagation of resource information.There is a growing interest in the interaction of Grid Computing and the P2P paradigm, pushing towards scalable solutions. This approach brings us to study the distribution information policies on P2P paradigm, to use them to compare their performance with other no distributed policies in a Grid Computing scenario.Previously is essential characterize the performance of some no distributed policies, hierarchical policiy for example.Previous works on hierarchical policy show that a random selection of root produces significally disimilar results.Finding a criteria for choosing the root node is essential to allow its usage in generated scenarios.The test of each possible root is not a scalable methodology and results very difficult to find a sufficiently reliable criteria.In the case of a three-level structure in topologies with the same node degree for all nodes (i.e. clique or circle), selecting any root produces no impact in the obtained results.For scale-free topologies, selecting the root becomes a nontrivial desicion.This work proposes a procedure to choose the root that obtains reasonably good performance for the hierarchical policiy in a wide range of analyzed scenarios.