Reeb graphs are a fundamental data structure for understanding and representing the topology of shapes. They used in computer graphics, solid modeling, visualization applications ranging from computation similarities finding defects complex models to automatic selection parameters.

Reeb graphs are a fundamental data structure for understanding and representing the topology of shapes. They used in computer graphics, solid modeling, visualization applications ranging from computation similarities finding defects complex models to automatic selection parameters. We introduce an on-line algorithm that reads stream elements (vertices, triangles, tetrahedra, etc.) continuously maintains graph all already reed. The is robust handling non-manifold meshes general its...

The IDX data format provides efficient, cache oblivious, and progressive access to large-scale scientific datasets by storing the in a hierarchical Z (HZ) order. Data stored can be visualized an interactive environment allowing for meaningful explorations with minimal resources. This technology enables real-time, visualization analysis of large on variety systems ranging from desktops laptop computers portable devices such as iPhones/iPads over web. While existing ViSUS API writing is...

This article presents a simple framework for progressive processing of high-resolution images with minimal resources. We demonstrate this framework's effectiveness by implementing an adaptive, multi-resolution solver gradient-based image that, the first time, is capable handling gigapixel imagery in real time. With our system, artists can use commodity hardware to interactively edit massive and apply complex operators, such as seamless cloning, panorama stitching, tone mapping. introduce...

Parallel I/O library performance can vary greatly in response to user-tunable parameter values such as aggregator count, file and aggregation strategy. Unfortunately, manual selection of these is time consuming dependent on characteristics the target machine, underlying system, dataset itself. Some characteristics, amount memory per core, also impose hard constraints range viable values. In this work we address problems by using machine learning techniques model PIDX parallel select...

In the era of big data, materials science workflows need to handle large-scale data distribution, storage, and computation. Any these areas can become a performance bottleneck. We present framework for analyzing internal material structures (e.g., cracks) mitigate bottlenecks. demonstrate effectiveness our workflow performing synchrotron X-ray computed tomography reconstruction segmentation silica-based structure. Our provides cloud-based, cutting-edge solution challenges such as growing...

Hierarchical, multiresolution data representations enable interactive analysis and visualization of large-scale simulations. One promising application these techniques is to store high performance computing simulation output in a hierarchical Z (HZ) ordering that translates from Cartesian coordinate scheme one-dimensional array ordered by locality at different resolution levels. However, when the dimensions are not an even power 2, parallel HZ produces sparse memory network access patterns...

We present a distributed framework that enables real-time streaming and visualization of data generated by large remote simulations. address issues arising from clientserver environments guarantee good parallel load balancing. apply progressive computing techniques parallel, hierarchical to reduce the "distance" between simulation hardware systems where actual analysis occur. simple efficient balancing method scales arbitrary sizes does not introduce additional communication cost....

We introduce a parallel approach to geometric modeling of complex objects and scenes, combining dataflow streaming BSP trees with partition the object space into independent portions, be evaluated in minimal interprocess communication. Binary Space Partition (BSP) is index used graphics for hidden-surface removal animation. use fuzzy leaves as progressive representation solid meshes. Our implemented processes that progress concurrently, where each refinement input process mapped instantly...

Hierarchical, multiresolution data representations enable interactive analysis and visualization of large-scale simulations. One promising application these techniques is to store high performance computing simulation output in a hierarchical Z (HZ) ordering that translates from Cartesian coordinate scheme one-dimensional array ordered by locality at different resolution levels. However, when the dimensions are not an even power 2, parallel HZ produces sparse memory network access patterns...

Hair designing is one of crucial components the hair simulation tasks.The efficiency modeling very much determined by interactivity and ease-to-use tools within an application.This paper presents a unified framework that uses various key techniques developed for specific tasks in to realize ultimate goal 'virtual hairdressing room' simple use but quite effective generating fast hairstyles.Successful attempts have been made handle different challenging issues involved at interactive...

This paper introduces a new progressive multi-resolution approach for representating and processing polyhedral objects of any dimension. Our representation, variant BSP trees [Nay90] combined with the Split scheme introduced in [BP96], allows streaming rendering solid models at multiple levels detail (LOD). Boolean set operations are computed progressively by reading input stream incremental refinements operands. Each refinement is mapped immediately to output so that result also represented...

As our ability to generate large and complex datasets grows, accessing processing these massive data collections is increasingly the primary bottleneck in scientific analysis. Challenges include retrieving, converting, resampling, combining remote often disparately located ensembles with only limited support from existing tools. In particular, solutions rely predominantly on extensive transfers or large-scale computing resources, both of which are inherently offline processes long delays...

Modern science is inundated with ever increasing data sizes as computational capabilities and image acquisition techniques continue to improve. For example, simulations are tackling larger domains higher fidelity, high-throughput microscopy generate that fundamental gather biologically medically relevant insights. As the exceed memory, even sometimes local disk space, each step in a scientific workflow impacted. Current software solutions enable exploration limited interactivity for...

In many areas of applied geometric/numeric computational mathematics, including geo-mapping, computer vision, graphics, finite element analysis, medical imaging, geometric design, and solid modeling, one has to compute incidences, adjacencies, ordering cells, generally using disparate often incompatible data structures algorithms. This article introduces topology algorithms discover the two-dimensional (2D)/3D space partition induced by a collection objects dimension 1D/2D, respectively....

This work presents NSDF-FUSE, a testbed for evaluating settings and performance of FUSE-based file systems on top S3-compatible object storage; the is part suite services from National Science Data Fabric (NSDF) project (an NSF-funded that delivering cyberinfrastructures data scientists). We demonstrate how NSDF-FUSE can be deployed to evaluate eight different mapping packages mount storage system, as well six patterns representing I/O operations two cloud platforms. open-source easily...

The lack of a readily accessible, tightly integrated data fabric connecting high-speed networking, storage, and computing services remains critical barrier to the democratization scientific discovery. To address this challenge, we are building National Science Data Fabric (NSDF), holistic ecosystem facilitate domain scientists in their daily research. NSDF comprises services, as well outreach initiatives. In paper, present testbed integrating three (i.e., computing). We evaluate performance....

The object of this paper is to transform a set local chain complexes single global complex using an equivalence relation congruence cells, solving topologically the numerical inaccuracies floating-point arithmetics. While computing space arrangement generated by collection cellular complexes, one may start from independently and efficiently intersection each input 2-cell with others. topology these intersections codified within (0-2)-dimensional complexes. target merge chains relations...

Computational resources are increasingly provisioned to users through cloud-like interfaces. Both academic and commercial cloud offerings exist, but no single standardized interface for common actions such as configuration, launching, termination of virtual exists. This imposes huge technical burden on domain scientist that attempt take advantage these resources; even expert spend considerable time port their applications from one platform another.