This work presents the I/O in-memory server implemented in the context of the Ophidia framework, a big data analyt-ics stack addressing scientific data analysis of n-dimensional datasets. The provided I/O server represents a key com-ponent in the Ophidia 2.0 architecture proposed in this pa-per. It exploits (i) a NoSQL approach to manage scientiffic data at the storage level, (ii) user-defined functions to per-form array-based analytics, (iii) the Ophidia Storage API to manage heterogeneous back-ends through a plugin-based approach, and (iv) an in-memory and parallel analytics en-gine to address high scalability and performance. Prelim-inary performance results about a statistical analytics ker-nel benchmark performed on a HPC cluster running at the CMCC SuperComputing Centre are provided in this paper.
An in-memory based framework for scientific data analytics / Elia, D.; Fiore, S.; D'Anca, A.; Palazzo, C.; Foster, I.; Williams, D. N.. - (2016), pp. 424-429. (Intervento presentato al convegno ACM International Conference on Computing Frontiers, CF 2016 tenutosi a ita nel 2016) [10.1145/2903150.2911719].
An in-memory based framework for scientific data analytics
Fiore S.;
2016-01-01
Abstract
This work presents the I/O in-memory server implemented in the context of the Ophidia framework, a big data analyt-ics stack addressing scientific data analysis of n-dimensional datasets. The provided I/O server represents a key com-ponent in the Ophidia 2.0 architecture proposed in this pa-per. It exploits (i) a NoSQL approach to manage scientiffic data at the storage level, (ii) user-defined functions to per-form array-based analytics, (iii) the Ophidia Storage API to manage heterogeneous back-ends through a plugin-based approach, and (iv) an in-memory and parallel analytics en-gine to address high scalability and performance. Prelim-inary performance results about a statistical analytics ker-nel benchmark performed on a HPC cluster running at the CMCC SuperComputing Centre are provided in this paper.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione
