Scientific Data Analysis Applications Research Articles

Evaluating Text Reading Speed in VR Scenes and 3D Particle Visualizations.

This work reports how text size and other rendering conditions affect reading speeds in a virtual reality environment and a scientific data analysis application. Displaying text legibly yet space-efficiently is a challenging problem in immersive displays. Effective text displays that enable users to read at their maximum speed must consider the variety of virtual reality (VR) display hardware and possible visual exploration tasks. We investigate how text size and display parameters affect reading speed and legibility in three state-of-the-art VR displays: two head-mounted displays and one CAVE. In our perception experiments, we establish limits where reading speed declines as the text size approaches the so-called critical print sizes (CPS) of individual displays, which can inform the design of uniform reading experiences across different VR systems. We observe an inverse correlation between display resolution and CPS. Yet, even in high-fidelity VR systems, the measured CPS was larger than in comparable physical text displays, highlighting the value of increased VR display resolutions in certain visualization scenarios. Our findings indicate that CPS can be an effective metric for evaluating VR display usability. Additionally, we evaluate the effects of text panel placement, orientation, and occlusion-reducing rendering methods on reading speeds in generic volumetric particle visualizations. Our study provides insights into the trade-off between text representation and legibility in cluttered immersive environments with specific suggestions for visualization designers and highlight areas for further research.

HEP-Frame: Improving the efficiency of pipelined data transformation & filtering for scientific analyses

HEP-Frame: Improving the efficiency of pipelined data transformation & filtering for scientific analyses

Post-processing for ITER Scrape-Off Layer Plasma Simulations (SOLPS-ITER) in IMAS Framework

With the ongoing effort to establish the Integrated Modelling and Analysis Suite (IMAS) and the Interface Data Structures (IDSs) as the standard relevant software infrastructure for ITER Organization and its parties for all ITER related activities, we report on steps towards the integration of the B2.5-Eirene plasma fluid solver, one of the codes in the SOLPS-ITER code package, within IMAS workflows, by means of a set of modules which allow post-processing of solution data into the IMAS model. This way, after a case is successfully completed, the solver simulation output results can be stored to the IDSs. To provide the means for visualization of the edge plasma state from the IDSs, a plugin for ParaView, an open-source scientific data analysis and visualization application, named ReadUALEdge was developed. With the data displayed within the ParaView application, all tools and utilities provided by ParaView can be used on the data for convenient and efficient analysis. In this article, we give an overview of the IMAS framework and SOLPS-ITER status and present the main aspects and features of the latest ReadUALEdge plugin for the ParaView application.

Accelerate Scientific Deep Learning Models on Heterogeneous Computing Platform with FPGA

AI and deep learning are experiencing explosive growth in almost every domain involving analysis of big data. Deep learning using Deep Neural Networks (DNNs) has shown great promise for such scientific data analysis applications. However, traditional CPU-based sequential computing without special instructions can no longer meet the requirements of mission-critical applications, which are compute-intensive and require low latency and high throughput. Heterogeneous computing (HGC), with CPUs integrated with GPUs, FPGAs, and other science-targeted accelerators, offers unique capabilities to accelerate DNNs. Collaborating researchers at SHREC1at the University of Florida, CERN Openlab, NERSC2at Lawrence Berkeley National Lab, Dell EMC, and Intel are studying the application of heterogeneous computing (HGC) to scientific problems using DNN models. This paper focuses on the use of FPGAs to accelerate the inferencing stage of the HGC workflow. We present case studies and results in inferencing state-of-the-art DNN models for scientific data analysis, using Intel distribution of OpenVINO, running on an Intel Programmable Acceleration Card (PAC) equipped with an Arria 10 GX FPGA. Using the Intel Deep Learning Acceleration (DLA) development suite to optimize existing FPGA primitives and develop new ones, we were able accelerate the scientific DNN models under study with a speedup from 2.46x to 9.59x for a single Arria 10 FPGA against a single core (single thread) of a server-class Skylake CPU.

An introduction to algorithmic differentiation

AbstractAlgorithmic differentiation (AD), also known as automatic differentiation, is a technology for accurate and efficient evaluation of derivatives of a function given as a computer model. The evaluations of such models are essential building blocks in numerous scientific computing and data analysis applications, including optimization, parameter identification, sensitivity analysis, uncertainty quantification, nonlinear equation solving, and integration of differential equations. We provide an introduction to AD and present its basic ideas and techniques, some of its most important results, the implementation paradigms it relies on, the connection it has to other domains including machine learning and parallel computing, and a few of the major open problems in the area. Topics we discuss include: forward mode and reverse mode of AD, higher‐order derivatives, operator overloading and source transformation, sparsity exploitation, checkpointing, cross‐country mode, and differentiating iterative processes.This article is categorized under: Algorithmic Development > Scalable Statistical Methods Technologies > Data Preprocessing

Parallel multi-dimensional range query processing with R-trees on GPU

The general purpose computing on graphics processing unit (GP-GPU) has emerged as a new cost effective parallel computing paradigm in high performance computing research that enables large amount of data to be processed in parallel. Large scale scientific data intensive applications have been playing an important role in modern high performance computing research. A common access pattern into such scientific data analysis applications is multi-dimensional range query, but not much research has been conducted on multi-dimensional range query on the GPU. Inherently multi-dimensional indexing trees such as R-Trees are not well suited for GPU environment because of its irregular tree traversal. Traversing irregular tree search path makes it hard to maximize the utilization of massively parallel architectures. In this paper, we propose a novel MPTS (Massively Parallel Three-phase Scanning) R-tree traversal algorithm for multi-dimensional range query, that converts recursive access to tree nodes into sequential access. Our extensive experimental study shows that MPTS R-tree traversal algorithm on NVIDIA Tesla M2090 GPU consistently outperforms traditional recursive R-trees search algorithm on Intel Xeon E5506 processors.

An Exploration of CUDA and CBEA for a Gravitational Wave Data-Analysis Application (Einstein@Home)

We present a detailed approach for making use of two new computer hardware architectures -- CBEA and CUDA -- for accelerating a scientific data-analysis application (Einstein@Home). Our results suggest that both the architectures suit the application quite well and the achievable performance in the same software developmental time-frame, is nearly identical.

The design and implementation of a workflow analysis tool

Motivated by the use of scientific workflows as a user-oriented mechanism for building executable scientific data integration and analysis applications, this article introduces a framework and a set of associated methods for analysing the execution properties of scientific workflows. Our framework uses a number of formal modelling techniques to characterize the process and data behaviour of workflows and workflow components and to reason about their functional and execution properties. We use the framework to design the architecture of a customizable tool that can be used to analyse the key execution properties of scientific workflows at authoring stage. Our design is generic and can be applied to a wide variety of scientific workflow languages and systems, and is evaluated by building a prototype of the tool for the Discovery Net system. We demonstrate and discuss the utility of the framework and tool using workflows from a real-world medical informatics study.

A Tool for Prioritizing DAGMan Jobs and its Evaluation

It is often difficult to perform efficiently a collection of jobs with complex job dependencies due to temporal unpredictability of the Grid. One way to mitigate the unpredictability is to schedule job execution in a manner that constantly maximizes the number of jobs that can be sent to workers. A recently developed scheduling theory provides a basis to meet that optimization goal. Intuitively, when the number of such jobs is always large, high parallelism can be maintained, even if the number of workers changes over time in an unpredictable manner. In this paper we present the design, implementation, and evaluation of a practical scheduling tool inspired by the theory. Given a DAGMan input file with interdependent jobs, the tool prioritizes the jobs. The resulting schedule significantly outperforms currently used schedules under a wide range of system parameters, as shown by simulation studies. For example, a scientific data analysis application, AIRSN, was executed at least 13% faster with 95% confidence. An implementation of the tool was integrated with the Condor high-throughput computing system.

Cataloging and Mining Massive Datasets for Science Data Analysis

With hardware advances in sensors, scientific instruments, and data storage techniques has come the inevitable flood of data that threatens to render traditional approaches to data analysis inadequate. The classic paradigm of a scientist manually and exhaustively going through a dataset is no longer feasible for many problems, ranging from remote sensing, astronomy, and atmospheric science to medicine, molecular biology, and biochemistry. This article presents our views as practitioners engaged in building computational systems to help scientists analyze and reduce massive datasets. We focus on what we view as challenges and shortcomings of the current state-of-the-art in data analysis in view of the massive datasets that are still awaiting analysis. The presentation focuses on recent and current scientific data analysis applications in astronomy, planetary sciences, solar physics, and atmospheric science that we have been involved with at the Jet Propulsion Laboratory (JPL).