In this paper, we propose a novel cost model for Spark SQL. The cost model covers the class of Generalized Projection, Selection, Join (GPSJ) queries. The cost model keeps into account the network and IO costs as well as the most relevant CPU costs. The execution cost is computed starting from a physical plan produced by Spark. The set of operations adopted by Spark when executing a GPSJ query are analytically modeled based on the cluster and application parameters, together with a set of database statistics. Experimental results carried out on three benchmarks and on two clusters of different sizes and with different computation features show that our model can estimate the actual execution time with about the 20 percent of errors on the average. Such an accuracy is good enough to let the system choose the most effective plan even when the execution time differences are limited. The error can be reduced to 14 percent, if the analytic model is coupled with our straggler handling strategy.

UNLABELLED: Extracting chemical features like Atom-Atom Mapping (AAM), Bond Changes (BCs) and Reaction Centres from biochemical reactions helps us understand the chemical composition of enzymatic reactions. Reaction Decoder is a robust command line tool, which performs this task with high accuracy. It supports standard chemical input/output exchange formats i.e. RXN/SMILES, computes AAM, highlights BCs and creates images of the mapped reaction. This aids in the analysis of metabolic pathways and the ability to perform comparative studies of chemical reactions based on these features. AVAILABILITY AND IMPLEMENTATION: This software is implemented in Java, supported on Windows, Linux and Mac OSX, and freely available at https://github.com/asad/ReactionDecoder CONTACT: : [email protected] or [email protected].

The emerging medical models aim at leveraging on high-throughput genome sequencing technologies to better target drugs to patients' personal profiles so as to increase their effectiveness. However, the huge amount of data made available by these technologies calls for sophisticated and automated analysis techniques. In this direction we present GOLAM, a framework for OLAP analysis and mining of matches between genomic regions extracted from ENCODE, a worldwide-available collection of shared genomic data. The goal of GOLAM is to overcome the current limitations of genome analysis methods, that are normally based on browsing. This is done by partially automating and speeding-up the analysis process on the one hand, by making it more flexible and introducing a multi-resolution view of data on the other. The framework has been partially implemented so far; in this paper we focus on conveying its potential and on describing its functional architecture and the underlying data models.