Gregory P. Johnson

DisplayCluster: An Interactive Visualization Environment for Tiled Displays

Display Cluster is an interactive visualization environment for cluster-driven tiled displays. It provides a dynamic, desktop-like windowing system with built-in media viewing capability that supports ultra high-resolution imagery and video content and streaming that allows arbitrary applications from remote sources (such as laptops or remote visualization machines) to be shown. This support extends to high-performance parallel visualization applications, enabling interactive streaming and display for hundred-mega pixel dynamic content. Display Cluster also supports multi-user, multi-modal interaction via devices such as joysticks, smart phones, and the Microsoft Kinect. Further, our environment provides a Python-based scripting interface to automate any set of interactions. In this paper, we describe the features and architecture of Display Cluster, compare it to existing tiled display environments, and present examples of how it can combine the capabilities of large-scale remote visualization clusters and high-resolution tiled display systems. In particular, we demonstrate that Display Cluster can stream and display up to 36 mega pixels in real time and as many as 144 mega pixels interactively, which is 3× faster and 4× larger than other available display environments. Further, we achieve over a gig pixel per second of aggregate bandwidth streaming between a remote visualization cluster and our tiled display system.

OSPRay - A CPU Ray Tracing Framework for Scientific Visualization

Scientific data is continually increasing in complexity, variety and size, making efficient visualization and specifically rendering an ongoing challenge. Traditional rasterization-based visualization approaches encounter performance and quality limitations, particularly in HPC environments without dedicated rendering hardware. In this paper, we present OSPRay, a turn-key CPU ray tracing framework oriented towards production-use scientific visualization which can utilize varying SIMD widths and multiple device backends found across diverse HPC resources. This framework provides a high-quality, efficient CPU-based solution for typical visualization workloads, which has already been integrated into several prevalent visualization packages. We show that this system delivers the performance, high-level API simplicity, and modular device support needed to provide a compelling new rendering framework for implementing efficient scientific visualization workflows.

CPU ray tracing large particle data with balanced P-k-d trees

We present a novel approach to rendering large particle data sets from molecular dynamics, astrophysics and other sources. We employ a new data structure adapted from the original balanced k-d tree, which allows for representation of data with trivial or no overhead. In the OSPRay visualization framework, we have developed an efficient CPU algorithm for traversing, classifying and ray tracing these data. Our approach is able to render up to billions of particles on a typical workstation, purely on the CPU, without any approximations or level-of-detail techniques, and optionally with attribute-based color mapping, dynamic range query, and advanced lighting models such as ambient occlusion and path tracing.

EnVision: A Web-Based Tool for Scientific Visualization

Scientific visualization is the process of transforming raw numeric data into a visual form, and is a key element of computational science. While many tools exist, they are unnecessarily difficult to use. This complexity increases time to insight and inhibits casual inquiry. The complexity derives from the need to support arbitrarily formatted data and many visualization algorithms. EnVision addresses both sources of complexity. Its design is predicated on two key insights. First, though the number of data file formats is unbounded, the structure of any one can be described using a small number of parameters. Second, the set of visualization algorithms applicable to a given type of data is small, and the subset used within a specific scientific discipline is smaller. EnVision utilizes domain-specific knowledge and user-directed semi-automation to dramatically simplify data importation and visualization algorithm selection. Its web-based interface facilitates access to remote hardware resources and provides a collaborative visualization environment.

Interactive Visualization and Analysis of Transitional Flow

A stand-alone visualization application has been developed by a multi-disciplinary, collaborative team with the sole purpose of creating an interactive exploration environment allowing turbulent flow researchers to experiment and validate hypotheses using visualization. This system has specific optimizations made in data management, caching computations, and visualization allowing for the interactive exploration of datasets on the order of 1TB in size. Using this application, the user (co-author Calo) is able to interactively visualize and analyze all regions of a transitional flow volume, including the laminar, transitional and fully turbulent regions. The underlying goal of the visualizations produced from these transitional flow simulations is to localize turbulent spots in the laminar region of the boundary layer, determine under which conditions they form, and follow their evolution. The initiation of turbulent spots, which ultimately lead to full turbulence, was located via a proposed feature detection condition and verified by experimental results. The conditions under which these turbulent spots form and coalesce are validated and presented.

Optimizing Distribution of Pandemic Influenza Antiviral Drugs

Effective distribution of these drugs will reduce illness and death in underinsured populations.

Configurable data prefetching scheme for interactive visualization of large-scale volume data

This paper presents a novel data prefetching and memory management scheme to support interactive visualization of large-scale volume datasets using GPU-based isosurface extraction. Our dynamic in-core approach uses a span-space lattice data structure to predict and prefetch the portions of a dataset that are required by isosurface queries, to manage an application-level volume data cache, and to ensure load-balancing for parallel execution. We also present a GPU memory management scheme that enhances isosurface extraction and rendering performance. With these techniques, we achieve rendering performance superior to other in-core algorithms while using dramatically fewer resources.

Stockpiling ventilators for influenza pandemics

In preparing for influenza pandemics, public health agencies stockpile critical medical resources. Determining appropriate quantities and locations for such resources can be challenging, given the considerable uncertainty in the timing and severity of future pandemics. We introduce a method for optimizing stockpiles of mechanical ventilators, which are critical for treating hospitalized influenza patients in respiratory failure. As a case study, we consider the US state of Texas during mild, moderate, and severe pandemics. Optimal allocations prioritize local over central storage, even though the latter can be deployed adaptively, on the basis of real-time needs. This prioritization stems from high geographic correlations and the slightly lower treatment success assumed for centrally stockpiled ventilators. We developed our model and analysis in collaboration with academic researchers and a state public health agency and incorporated it into a Web-based decision-support tool for pandemic preparedness and response.

Visualizing Turbulent Flow

Introduction The images shown in Figure 1 display a single time step of a turbulent flow simulation computed by Dr. Thomas J. R. Hughes, Professor of Aerospace Engineering, and Victor Calo at The University of Texas at Austin. This research examines how a fluid running over a flat plate suddenly becomes turbulent. The smooth flow seen at the left-hand side of the volume has low-amplitude fluctuations that interact with the smooth, laminar boundary layer of the fluid. As the energy of the fluctuations is convected downward into the boundary layer, there is a sudden explosion of the entire flow into turbulence. The left-most image of Figure 1 displays the left to right nature of the fluid as it passes over the flat plate, showing the turbulent boundary layer. The middle image shows the volumetric flow with sheet of particles inserted at designated locations to show increasing turbulent behavior. The right-most image in Figure 1 is a close up of particle sheets showing local velocity fluctuations inside the turbulent boundary layer.

Equalizing access to pandemic influenza vaccines through optimal allocation to public health distribution points

Vaccines are arguably the most important means of pandemic influenza mitigation. However, as during the 2009 H1N1 pandemic, mass immunization with an effective vaccine may not begin until a pandemic is well underway. In the U.S., state-level public health agencies are responsible for quickly and fairly allocating vaccines as they become available to populations prioritized to receive vaccines. Allocation decisions can be ethically and logistically complex, given several vaccine types in limited and uncertain supply and given competing priority groups with distinct risk profiles and vaccine acceptabilities. We introduce a model for optimizing statewide allocation of multiple vaccine types to multiple priority groups, maximizing equal access. We assume a large fraction of available vaccines are distributed to healthcare providers based on their requests, and then optimize county-level allocation of the remaining doses to achieve equity. We have applied the model to the state of Texas, and incorporated it in a Web-based decision-support tool for the Texas Department of State Health Services (DSHS). Based on vaccine quantities delivered to registered healthcare providers in response to their requests during the 2009 H1N1 pandemic, we find that a relatively small cache of discretionary doses (DSHS reserved 6.8% in 2009) suffices to achieve equity across all counties in Texas.