Browsing by Author "Gupta, Anshul"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    Design and Implementation of a Scalable Parallel Direct Solver for Sparse Symmetric Positive Definite Systems: Preliminary Results
    (1997) Gupta, Anshul; Gustavson, Fred; Joshi, Mahesh; Karypis, George; Kumar, Vipin
    Solving large sparse systems of linear equations is at the core of many problems in engineering and scientific computing. It has long been a challenge to develop parallel formulations of sparse direct solvers due to several different complex steps involved in the process. In this paper, we describe one of the first efficient, practical, and robust parallel solvers for sparse symmetric positive definite linear systems that we have developed and discuss the algorithmic and implementation issues involved in its development.
  • Thumbnail Image
    Item
    Head Mounted Eye Tracking Aid for Central Visual Field Loss
    (2016-07) Gupta, Anshul
    Age-Related Macular Degeneration results in central visual field loss (CFL) due to formation of central blind-spots or scotomas. Activities like reading are affected. We hypothesize that real-time remapping of lost information due to CFL onto a functional portion of the retina will improve visual performance. We have developed two hardware prototypes using a head-mounted display, integrated eye-tracker, and computer to remap and display images in real-time to the wearer. To test, in three different studies, normally-sighted subjects were asked to wear the head-mounted display with the built-in eye tracker. CFL was simulated by placing artificial circular scotomas ranging from 2° to 16° diameter over the gaze position, and reading speed was measured for the remapped and unremapped condition. We observed a statistically significant increase in mean reading speeds for the larger scotomas. Results indicate that the device shows promise for use with CFL patients.
  • Thumbnail Image
    Item
    Parallel Algorithm Scalability Issues in PetaFLOPS Architectures
    (2001-01-26) Garma, Ananth; Gupta, Anshul; Han, Euihong; Kumar, Vipin
    The projected design space of petaFLOPS architectures entails exploitationof very large degrees of concurrency, locality of data access, and toleranceto latency. This puts considerable pressure on the design of parallelalgorithms capable of effectively utilizing increasing amounts of processingresources in a memory and bandwidth constrained environment. This aspect ofalgorithm design, also referred to as scalability analysis, is a keycomponent for guiding algorithm designers as well as hardware architects.By identifying bottlenecks to scalability and machine parameters thatinfluence these bottlenecks, scalability analysis provides insights toalleviating the bottlenecks in the context of the specific algorithm.In this paper, we motivate the need for, and benefits of scalabilityanalysis in the context of petaFLOPS systems. We overview variousscalability metrics and study their suitability to petaFLOPS system.We also present sample analysis of selected computational kernels fromdense linear algebra, fast fourier transforms, and data intensive applications(association rule mining). The objective of this analysis is to demonstratethe analysis framework and its use in identifying desirable architecturalfeatures as well the ability of these selected kernels to scale to petaFLOPSsystems.