Search results for "Large-Scale"
showing 3 items of 83 documents
Measuring galaxy segregation with the mark connection function
2010
(abridged) The clustering properties of galaxies belonging to different luminosity ranges or having different morphological types are different. These characteristics or `marks' permit to understand the galaxy catalogs that carry all this information as realizations of marked point processes. Many attempts have been presented to quantify the dependence of the clustering of galaxies on their inner properties. The present paper summarizes methods on spatial marked statistics used in cosmology to disentangle luminosity, colour or morphological segregation and introduces a new one in this context, the mark connection function. The methods used here are the partial correlation functions, includi…
The meaningfulness of two curriculum-based national tests of English as a foreign language
2022
This paper addresses the aspect of the meaningfulness of a national assessment in English as a foreign language, applying the fairness framework proposed by Kunnan (2018). We compared students’ performance on two receptive language skills of listening and reading on two subsequent national evaluations in Finland, taken by students at the end of compulsory basic education and at the end of general upper secondary education, respectively. The research questions focus on (1) the relationship between students’ performance on the two tests and their gender, language of schooling and parents’ educational level, and (2) the relationship between the students’ receptive language proficiency at the e…
Designing a graphics processing unit accelerated petaflop capable lattice Boltzmann solver: Read aligned data layouts and asynchronous communication
2016
The lattice Boltzmann method is a well-established numerical approach for complex fluid flow simulations. Recently, general-purpose graphics processing units (GPUs) have become available as high-performance computing resources at large scale. We report on designing and implementing a lattice Boltzmann solver for multi-GPU systems that achieves 1.79 PFLOPS performance on 16,384 GPUs. To achieve this performance, we introduce a GPU compatible version of the so-called bundle data layout and eliminate the halo sites in order to improve data access alignment. Furthermore, we make use of the possibility to overlap data transfer between the host central processing unit and the device GPU with com…