Influenced by the advances in data and computing, the scientific practice increasingly involves machine learning and artificial intelligence driven methods which requires specialized capabilities at the system-, science- and service-level in addition to the conventional large-capacity supercomputing approaches. The latest distributed architectures built around the composability of data-centric applications led to the emergence of a new ecosystem for container coordination and integration. However, there is still a divide between the application development pipelines of existing supercomputing environments, and these new dynamic environments that disaggregate fluid resource pools through accessible, portable and re-programmable interfaces. New approaches for dynamic composability of heterogeneous systems are needed to further advance the data-driven scientific practice for the purpose of more efficient computing and usable tools for specific scientific domains. In this paper, we present a novel approach for using composable systems in the intersection between scientific computing, artificial intelligence (AI), and remote sensing domain. We describe the architecture of a first working example of a composable infrastructure that federates Expanse, an NSF-funded supercomputer, with Nautilus, a Kubernetes-based GPU geo-distributed cluster. We also summarize a case study in wildfire modeling, that demonstrates the application of this new infrastructure in scientific workflows: a composed system that bridges the insights from edge sensing, AI and computing capabilities with a physics-driven simulation.

HTCondor is a major workload management system used in distributed high throughput computing (dHTC) environments, e.g., the Open Science Grid. One of the distinguishing features of HTCondor is the native support for data movement, allowing it to operate without a shared filesystem. Coupling data handling and compute scheduling is both convenient for users and allows for significant infrastructure flexibility but does introduce some limitations. The default HTCondor data transfer mechanism routes both the input and output data through the submission node, making it a potential bottleneck. In this document we show that by using a node equipped with a 100 Gbps network interface (NIC) HTCondor can serve data at up to 90 Gbps, which is sufficient for most current use cases, as it would saturate the border network links of most research universities at the time of writing.

In 2016, a team of earth scientists directly engaged a team of computer scientists to identify cyberinfrastructure (CI) approaches that would speed up an earth science workflow. This paper describes the evolution of that workflow as the two teams bridged CI and an image segmentation algorithm to do large scale earth science research. The Pacific Research Platform (PRP) and The Cognitive Hardware and Software Ecosystem Community Infrastructure (CHASE-CI) resources were used to significantly decreased the earth science workflow's wall-clock time from 19.5 days to 53 minutes. The improvement in wall-clock time comes from the use of network appliances, improved image segmentation, deployment of a containerized workflow, and the increase in CI experience and training for the earth scientists. This paper presents a description of the evolving innovations used to improve the workflow, bottlenecks identified within each workflow version, and improvements made within each version of the workflow, over a three-year time period.

High resolution satellite imagery is a growing source of data with potential applications in many diverse domains. Efficient large scale analysis of this rich data can lead to unprecedented discoveries with societal impact. We present a new framework for organizing collections of satellite images into demographically relevant categories using unsupervised learning techniques. Our framework first extracts features using pre-trained Convolutional Neural Networks from tiles of high resolution satellite images of a city. The k-means algorithm is then applied to these features to organize images into visually similar groups. The resulting clustered images are validated using demographic data. The cluster model is then applied to six different cities around the world to test the transferability of our methods. Finally, the discovered image clusters are visualized in our customized web interface to enable demographers, social scientists, and economists to understand the organization of a city.

Summary form only given. In this virtual-reality art installation - a mathematical play space - the viewer-participant creates an immersive visual and sonic experience. It is based on the mathematical and physical simulation of over one million particles with momentum and elastic reflection in an environment with gravity. The final scene has a realistic rendering of water with reflections and lighting based on spherical harmonics. Sound components are triggered and modified by the user and particle interaction. The application was originally developed using a CUDA particle system running within Thumb, a virtual-reality framework developed by Robert Kooima. It is now being ported to CalVR, developed by researchers at the California Institute for Telecommunications and Information Technology (Calit2) Qualcomm Institute at University of California, San Diego.

A modern field science such as archaeology is heavily data-driven using various kinds of state-of-the-art measurement instruments. It requires sophisticated computer infrastructure to manage large amounts of heterogeneous data. The concept of cloud computing provides a flexible cyber infrastructure for large-scale data management, which is being deployed at university campuses. A problem unique to field research is that researchers often work at remote field sites with limited computer and network resources. For a data management system that has to work in the campus cloud and under vastly different field conditions, portability of computer infrastructure and common data access methods are essential requirements. This paper explores the portability of cloud infrastructure and illustrates the portable data management system that we used in a recent archaeological expedition.

The broad adoption of diagnostic and analytical techniques in the field of archaeology, presents a unique opportunity for e-Science in the form of scientific explanation, drawing from methodologies aimed at recording, archiving, analyzing, and disseminating, rich data collections. This paper presents a needed stepping stone towards cyber-archaeology, creating synergy between information technology and archeology, by introducing a data acquisition, tagging and characterization pipeline.

Recognizing the value of autostereoscopy for 3D displays in public contexts, we pursue the goal of large-scale, high-resolution, immersive virtual reality using lenticular displays. Our contributions include the scalable tiling of lenticular displays to large fields of view and the use of GPU image interleaving and application optimization for real-time performance. In this context, we examine several ways to improve group-viewing by combining user tracking with multi-view displays.

Streaming very-high-definition visualization data objects on top of optical networks is critical in many scientific research areas, including video streaming/conferencing, remote rendering on tiled display walls, 3D virtual reality applications, etc. Current data streaming protocols rely on UDP as well as a variety of compression techniques. However, none of the protocols scale well to the parallel streaming model of large scale graphic applications, and the existing parallel streaming protocols have limited synchronization mechanisms to synchronize the streams efficiently, and are prone to be slowed down by just one slow stream. In this paper, we propose a new parallel streaming protocol that can stream synchronized multiple Gbps media content over optical networks through reliable Cross-Stream packet coding, which not only tolerates random UDP packet loss, but also aims to achieve good synchronization performance across multiple parallel data streams with reasonable coding overhead. We simulated the approach, and the results show that our approach can generate steady throughput with fluctuating data streams.

Many interrelated planetary height map and surface image map data sets exist, and more data are collected each day. Broad communities of scientists require tools to compose these data interactively and explore them via real-time visualization. While related, these data sets are often unregistered with one another, having different projection, resolution, format, and type. We present a GPU-centric approach to the real-time composition and display of unregistered-but-related planetary-scale data. This approach employs a GPGPU process to tessellate spherical height fields. It uses a render-to-vertex-buffer technique to operate upon polygonal surface meshes in image space, allowing geometry processes to be expressed in terms of image processing. With height and surface map data processing unified in this fashion, a number of powerful composition operations may be uniformly applied to both. Examples include adaptation to nonuniform sampling due to projection, seamless blending of data of disparate resolution or transformation regardless of boundary, and the smooth interpolation of levels of detail in both geometry and imagery. Issues of scalability and precision are addressed, giving out-of-core access to giga-pixel data sources, and correct rendering at scales approaching one meter.

A solid-state dynamic parallax barrier autostereoscopic display mitigates some of the restrictions present in static barrier systems, such as fixed view-distance range, slow response to head movements, and fixed stereo operating mode. By dynamically varying barrier parameters in real time, viewers may move closer to the display and move faster laterally than with a static barrier system, and the display can switch between 3D and 2D modes by disabling the barrier on a per-pixel basis. Moreover, Dynallax can output four independent eye channels when two viewers are present, and both head-tracked viewers receive an independent pair of left-eye and right-eye perspective views based on their position in 3D space. The display device is constructed by using a dual-stacked LCD monitor where a dynamic barrier is rendered on the front display and a modulated virtual environment composed of two or four channels is rendered on the rear display. Dynallax was recently demonstrated in a small-scale head-tracked prototype system. This paper summarizes the concepts presented earlier, extends the discussion of various topics, and presents recent improvements to the system.

Autostereoscopic displays enable unencumbered immersive virtual reality, but at a significant computational expense. This expense impacts the feasibility of autostereo displays in high-performance real-time interactive applications. A new autostereo rendering algorithm named autostereo combiner addresses this problem using the programmable vertex and fragment pipelines of modern graphics processing units (GPUs). This algorithm is applied to the Varrier, a large-scale, head-tracked, parallax barrier autostereo virtual reality platform. In this capacity, the Combiner algorithm has shown performance gains of 4x over traditional parallax barrier rendering algorithms. It has enabled high-performance rendering at sub-pixel scales, affording a 2x increase in resolution and showing a 1.4x improvement in visual acuity

In this paper, the effects of three visual factors: scene complexity, stereovision and motion parallax on correct perception of a virtual object's size were analyzed in an immersive virtual environment. We designed a controlled experiments set to incorporate visual conditions that reflected all twelve different configuration combinations of the three visual factors. Under each visual condition, subject performed the task of making judgments of the sizes of a virtual object displayed at five different distances from him/her. A total number of eighteen subjects participated in our study. The subjects' judgments and the corresponding actual sizes of the virtual object were recorded. Based on the collected data, two quantitative measures of subjects' performance were derived and analyzed. The results of our experiments were consistent across the majority of the subject population and suggested that scene complexity and stereovision could have significant impact on the performance of a user of virtual environments to make correct judgments on a virtual object's size. On the contrary, motion parallax, either produced by the virtual environment or by the observer, might not be a significant factor in determining that performance

A novel barrier strip autostereoscopic (AS) display is demonstrated using a solid-state dynamic parallax barrier. A dynamic barrier mitigates restrictions inherent in static barrier systems such as fixed view distance range, slow response to head movements, and fixed stereo operating mode. By dynamically varying barrier parameters in real time, viewers may move closer to the display and move faster laterally than with a static barrier system. Furthermore, users can switch between 3D and 2D modes by disabling the barrier. Dynallax is head-tracked, directing view channels to positions in space reported by a tracking system in real time. Such head-tracked parallax barrier systems have traditionally supported only a single viewer, but by varying the barrier period to eliminate conflicts between viewers, Dynallax presents four independent eye channels when two viewers are present. Each viewer receives an independent pair of left and right eye perspective views based on their position in 3D space. The display device is constructed using a dual-stacked LCD monitor where a dynamic barrier is rendered on the front display and the rear display produces a modulated VR scene composed of two or four channels. A small-scale head-tracked prototype VR system is demonstrated.

Summary form only given. Recent availability of production-quality switched gigabit and 10 gigabit networks on an international scale has facilitated the emergence of extreme multimedia devices and applications. Digital cinema at four times HDTV resolution (4096 /spl times/ 2160) can now be streamed from Japan 15,000 kilometers over CineGrid gigabit networks. 100-megapixel tiled displays can be managed as large landscapes of information fed by up to 20 gigabits/second of data from Amsterdam. A 3D point-to-point life-size teleconferencing with no glasses or tracking encumbrances is eerily natural. Stereo HDTV from underwater reefs in Taiwan or robots at play in Korea, and live HDTV sent from smoker vents two miles under the surface of the ocean triply illustrate the extended reach of multimedia today. To protect content from disruption or theft, new prototype technology showed switches in Ottawa, Amsterdam, and Chicago transmitting twelve 800 megabit/second multimedia streams to San Diego with 256-bit AES encryption, in real-time, without any knowledge by the application. This keynote presents multimedia pushed to demonstrated limits at the extreme September iGrid2005 workshopat which 49 networked demonstrations from 20 countries took advantage of 100 gigabits of networking into San Diego from around the world, displayed on huge screens in the just-opened Calit2 building at UCSD. Credits and references are amply provided.

Virtual reality (VR) has long been hampered by the gear needed to make the experience possible; specifically, stereo glasses and tracking devices. Autostereoscopic display devices are gaining popularity by freeing the user from stereo glasses, however few qualify as VR displays. The Electronic Visualization Laboratory (EVL) at the University of Illinois at Chicago (UIC) has designed and produced a large scale, high resolution head-tracked barrier-strip autostereoscopic display system that produces a VR immersive experience without requiring the user to wear any encumbrances. The resulting system, called Varrier, is a passive parallax barrier 35-panel tiled display that produces a wide field of view, head-tracked VR experience. This paper presents background material related to parallax barrier autostereoscopy, provides system configuration and construction details, examines Varrier interleaving algorithms used to produce the stereo images, introduces calibration and testing, and discusses the camera-based tracking subsystem.

Emerging lambda-Grid systems employ wavelength-routed network with optical switches to enable dynamic on-demand lightpaths with multi-gigabit rate bandwidth to interconnect shared computing clusters of user domains. User-centric lambda-Grid systems enable user domains to act as distributed connectivity providers of shared wavelength resources; and as distributed network control service providers of brokering shared wavelength resources and of provisioning scheduled lightpaths. To support user-centric lambda-Grid systems, the proposed Dynamic User-centric Switched Optical Network (DUSON) architecture employs an Intelligent Control Service Proxy (ICSP) at each user domain to support wavelength connectivity brokerage and lightpath provisioning/restoration middleware for user applications to control on-demand lightpaths. Furthermore, ICSP employs the proposed Robust Fast Optical Reservation Protocol (RFORP) during lightpath provisioning/restoration to minimize wavelength reservation blocking and delay via localized rerouting and parallel concurrent wavelength reservation processing respectively.

High speed bulk data transfer is an important part of many data-intensive scientific applications. This paper describes an aggressive bulk data transfer scheme, called Reliable Blast UDP (RBUDP), intended for extremely high bandwidth, dedicated- or Quality-of-Service-enabled networks, such as optically switched networks. This paper also provides an analytical model to predict RBUDP's performance and compares the results of our model against our implementation of RBUDP. Our results show that RBUDP performs extremely efficiently over high speed dedicated networks and our model is able to provide good estimates of its performance.

A review of existing systems for performing manufacturing tasks in a collaborative virtual environment highlights the limitations for modeling object behavioral characteristics. It is difficult for users to describe the tasks to be performed in the environment. A behavioral layer architecture is presented to address these limitations. The layer consists of a virtual manufacturing lattice (VML) structure, an object library, and a virtual manufacturing script (VMS). The VML is a hierarchical structure to encapsulate object behavioral characteristics, which augments the scenegraph structure. The encapsulation is achieved by using a four-tuple structure (C,R,T,E), where C is the composition structure of the node which contains geometry, material, texture, and location data, R is the precedence relationship element, T is the trajectory relationship element, and E is the event control list used to describe the state information about the object. The VMS is implemented as a script and a parser that serves as an interface between the user and collaborative virtual reality software. The script is a grammar-like structure to describe manufacturing tasks. In the behavioral layer architecture, the script is parsed, and calls are issued to the underlying functions to initiate the tasks and to manipulate the lattice structure to perform the specified tasks. Availability of network bandwidth is a key ingredient for using our system for telecollaboration. As part of an example to demonstrate the implementation of VML-VMS, a study of network latency and saturation is performed for different network medium, packet sizes, sampling rate, and network protocols to explore ways of reducing network latency and display frame jitter.

Despite their attractive properties, networked virtual environments (net-VEs) are notoriously difficult to design, implement, and test due to the concurrency, real-time and networking features in these systems. Net-VEs demand high quality-of-service (QoS) requirements on the network to maintain natural and real-time interactions among users. The current practice for net-VE design is basically trial and error, empirical, and totally lacks formal methods. This paper proposes to apply a Petri net formal modeling technique to a net-VE-NICE (narrative immersive constructionist/collaborative environment), predict the net-VE performance based on simulation, and improve the net-VE performance. NICE is essentially a network of collaborative virtual reality systems called the CAVE-(CAVE automatic virtual environment). First, we introduce extended fuzzy-timing Petri net (EFTN) modeling and analysis techniques. Then, we present EFTN models of the CAVE, NICE, and transport layer protocol used in NICE: transmission control protocol (TCP). We show the possibility analysis based on the EFTN model for the CAVE. Then, by using these models and design/CPN as the simulation tool, we conducted various simulations to study real-time behavior, network effects and performance (latencies and jitters) of NICE. Our simulation results are consistent with experimental data.

This paper proposes a methodology for employing collaborative, immersive virtual environments as a high-end visualization interface for massive data-sets. The methodology employs feature detection, partitioning, summarization and decimation to significantly cull massive data-sets. These reduced data-sets are then distributed to the remote CAVEs, ImmersaDesks and desktop workstations for viewing. The paper also discusses novel techniques for collaborative visualization and meta-data creation.

The paper discusses a virtual mail system. V-mail is designed to be used in the CAVE virtual environment. The participants' mail messages and the ongoing modifications of the VE are maintained by a central server. V-mail's user interface is embodied in a virtual friend or pet that follows the participant as he/she interacts with the VE.

This paper presents an overview of the tele-immersion applications that have been built by collaborators around the world using the CAVERNsoft toolkit, and the lessons learned from building these applications. In particular the lessons learned are presented as a set of rules-of-thumb for developing tele-immersive applications in general.

The Electronic Visualization Laboratory (EVL) at the University of Illinois at Chicago (UIC) has partnered with dozens of computational scientists and engineers to create visualization and virtual reality (VR) devices and applications for collaborative exploration of scientific and engineering data. Since 1995, our research and development activities have incorporated emerging high bandwidth networks like the vBNS and the Internet2 in an effort now called Tele-Immersion. As a result of our six years' experience in building first and second generation VR devices to support these applications, we consider third generation VR devices that will provide desktop/office sized displays. Since no current technology is yet configurable with ideal resolution and size, we will first simulate these devices with available parts, and then build more advanced prototypes. We believe that the devices we propose to build using the new display technologies form a set of desirable human/computer interface requirements for successful Tele-Immersion adoption. A goal of this research is to develop clearly compelling prototypes so that these devices can be improved and reproduced by the private sector.

In most virtual reality applications, 3-d space is a passive, ambient continuum in which the objects of study are placed. When the 3-d space itself is the object of study, as with mathematical manifolds, VR is especially important as a visualization medium. We describe the visualization of such spaces in the CAVE virtual environment.

Object instancing is the efficient method of representing an hierarchical object with a directed graph instead of a tree. If this graph contains a cycle then the object it represents is a linear fractal. Linear fractals are difficult to render for three specific reasons: (1) ray-fractal intersection is not trivial, (2) surface normals are undefined and (3) the object aliases at all sampling resolutions.Ray-fractal intersections are efficiently approximated to sub-pixel accuracy using procedural bounding volumes and a careful determination of the size of a pixel, giving the perception that the surface is infinitely detailed. Furthermore, a surface normal for these non-differentiable surfaces is defined and analyzed. Finally, the concept of antialiasing "covers" is adapted and used to solve the problem of sampling fractal surfaces.An initial bounding volume estimation method is also described, allowing a linear fractal to be rendered given only its iterated, function system. A parallel implementation of these methods is described and applications of these results to the rendering of other fractal models are given.

No abstract available.

Realism in computer graphics typically requires using 24 or more bits/pixel to generate an image. This paper describes a method developed by the authors called "Color Cell Compression" or "CCC" that preserves at least a limited animation and local update capability yet yields extraordinary-looking color images in approximately two bits/pixel independent of image complexity. Three intermediate methods of compressing images to six, four and three bits/pixel respectively are also described. The CCC encoding process for a 640 × 480 image averages 11 seconds on a VAX 11/750, however, the CCC method does permit real-time decoding of these images using software look-up tables and conventional display hardware. The three intermediate methods may also be decoded in real time but have the added advantage of requiring only 3-4 seconds for encoding on a VAX 11/750.

Home computer users are just now discovering computer graphics. Modest extensions to BASIC allow plotting but not much more. The Bally Home Library Computer, however, has hardware to aid implementation of video games. Custom integrated circuits working on a 160×102 pixel (2 bits per pixel) color television screen allow certain forms of animation in real time. To give this power to the user, BASIC Zgrass has been designed and implemented. It is an extension of BASIC that allows parallel processes, picture objects that move, scale and group together as well as several drawing modes. There are also software controls of a three-voice music synthesizer, interactive input devices, a film camera and an IEEE bus interface. We will concentrate mainly on the language design for making it all easy to learn and use.

Recent trends in the manufacturing and sales of home computers indicate that a new form of instruction must be developed for teaching programming in the home. Students learn programming from interacting with computers and other students in a classroom setting. The home user may not have this option. Manuals do not teach; they are good for looking up information you have forgotten. Workbooks and textbooks do not have a structure conducive to teaching—you have to understand everything before you can figure out “undiagnosible syntax error” or “OC4.” An approach has been developed at UICC which uses sophisticated error handling and modest program verification to supervise the beginning user while he/she is constructing and executing computer graphics programs. New, inexperienced computer users are now doing interesting (to them) programs within a half hour. This “interactive workbook” concept will be expanded and is currently a working prototype of a teaching package to be available on a home computer graphics/color TV system next year.

Civilization and its requisite overhead have neatly brought us away from real-time interactions. The high priests of technology use unwieldly systems to perpetuate cybercrud--the art of using computers to put things over on people. This mentality can be countered by bringing to people systems that are easily learned and used--'habitable' systems. 'Habitability', then, is the quality of a system that makes it comfortable to learn and use.Computer graphics, is especially in a media production environment, must be highly people-oriented. This paper attempts to develop the two essential components for a people-oriented computer system--habitability and environment, using the educational media production laboratory known as the Circle Graphics Habitat as the example.