Cloud computing allows on-demand access to cheap computing resources. This capability can be used for solving problems autonomously by complete enumeration. We present here SDVRP a data structure based on range partitioning that allows to autonomously divide the computing tasks to as many nodes as are needed to meet a user-imposed dead-line (in the order of minutes) despite heterogeneity of nodes. The data structure monitors itself to deal with failures and changes in node capacities. We use simulation for a proof-of-concept of this data structure.
CITATION STYLE
Grampone, S., Litwin, W., & Schwarz, T. S. J. (2015). A dependable, scalable, distributed, virtual data structure. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9532, pp. 724–735). Springer Verlag. https://doi.org/10.1007/978-3-319-27161-3_66
Mendeley helps you to discover research relevant for your work.