skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: PANDORA: a parallelizing approximation-discovery framework (WIP paper)
In this paper, we introduce PANDORA---a framework that complements existing parallelizing compilers by automatically discovering application- and architecture-specialized approximations. We demonstrate that PANDORA creates approximations that extract massive amounts of parallelism from inherently sequential code by eliminating loop-carried dependencies---a long-time goal of the compiler research community. Compared to exact parallel baselines, preliminary results show speedups ranging from 2.3x to 81x with acceptable error for many usage scenarios.  more » « less
Award ID(s):
1718033
PAR ID:
10104075
Author(s) / Creator(s):
;
Date Published:
Journal Name:
Proceedings of the 20th ACM SIGPLAN/SIGBED International Conference on Languages, Compilers, and Tools for Embedded Systems
Page Range / eLocation ID:
198 to 202
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; (, ACM Transactions on Embedded Computing Systems)
    null (Ed.)
    In this article, we introduce a p arallelizing a pproximatio n - d isc o very f ra mework, PANDORA, for automatically discovering application- and architecture-specialized approximations of provided code. PANDORA complements existing compilers and runtime optimizers by generating approximations with a range of Pareto-optimal tradeoffs between performance and error, which enables adaptation to different inputs, different user preferences, and different runtime conditions (e.g., battery life). We demonstrate that PANDORA can create parallel approximations of inherently sequential code by discovering alternative implementations that eliminate loop-carried dependencies. For a variety of functions with loop-carried dependencies, PANDORA generates approximations that achieve speedups ranging from 2.3x to 81x, with acceptable error for many usage scenarios. We also demonstrate PANDORA’s architecture-specialized approximations via FPGA experiments, and highlight PANDORA’s discovery capabilities by removing loop-carried dependencies from a recurrence relation with no known closed-form solution. 
    more » « less
  2. ; ; ; ; ; ; ; ; ; (, Atmospheric Chemistry and Physics)
    Abstract. Accurate representation of the hourly variation in the NO2-column-to-surface relationship is essential for interpreting geostationary observations of NO2 columns. Previous research indicated inconsistencies in this hourly variation. This study employs the high-performance configuration of the GEOS-Chem model (GCHP) to analyze daytime hourly NO2 total columns and surface concentrations during summer. We use measurements from globally distributed Pandora sun photometers and aircraft observations over the United States. We correct Pandora total NO2 vertical columns for (1) hourly variations in effective temperature driven by vertically resolved contributions to the total column and (2) changes in local solar time along the Pandora line of sight. These corrections increase the total NO2 columns by 5–6 × 1014 molec. cm−2 at 09:00 and 18:00 across all sites. Fine-scale simulations from GHCP (∼12 km) reduce the normalized bias (NB) against Pandora total NO2 columns from 19 % to 10 % and against aircraft measurements from 25 % to 13 % in Maryland, Texas, and Colorado. Similar reductions are observed in NO2 columns over the eastern US (17 % to 9 %), the western US (22 % to 14 %), Europe (24 % to 15 %), and Asia (29 % to 21 %) when compared to 55 km simulations. Our analysis attributes the weaker hourly variability in the total NO2 column to (1) hourly variations in column effective temperature, (2) local solar time changes along the Pandora line of sight, and (3) differences in hourly NO2 variability from different atmospheric layers, with the lowest 500 m exhibiting greater variability, while the dominant residual column above 500 m exhibits weaker variability. 
    more » « less
  3. ; ; ; ; ; ; ; ; ; ; et al (, The European Physical Journal C)
    Abstract The Pandora Software Development Kit and algorithm libraries perform reconstruction of neutrino interactions in liquid argon time projection chamber detectors. Pandora is the primary event reconstruction software used at the Deep Underground Neutrino Experiment, which will operate four large-scale liquid argon time projection chambers at the far detector site in South Dakota, producing high-resolution images of charged particles emerging from neutrino interactions. While these high-resolution images provide excellent opportunities for physics, the complex topologies require sophisticated pattern recognition capabilities to interpret signals from the detectors as physically meaningful objects that form the inputs to physics analyses. A critical component is the identification of the neutrino interaction vertex. Subsequent reconstruction algorithms use this location to identify the individual primary particles and ensure they each result in a separate reconstructed particle. A new vertex-finding procedure described in this article integrates a U-ResNet neural network performing hit-level classification into the multi-algorithm approach used by Pandora to identify the neutrino interaction vertex. The machine learning solution is seamlessly integrated into a chain of pattern-recognition algorithms. The technique substantially outperforms the previous BDT-based solution, with a more than 20% increase in the efficiency of sub-1 cm vertex reconstruction across all neutrino flavours. 
    more » « less
  4. ; ; ; ; ; ; ; ; ; ; et al (, The European Physical Journal C)
    Abstract The Pandora Software Development Kit and algorithm libraries provide pattern-recognition logic essential to the reconstruction of particle interactions in liquid argon time projection chamber detectors. Pandora is the primary event reconstruction software used at ProtoDUNE-SP, a prototype for the Deep Underground Neutrino Experiment far detector. ProtoDUNE-SP, located at CERN, is exposed to a charged-particle test beam. This paper gives an overview of the Pandora reconstruction algorithms and how they have been tailored for use at ProtoDUNE-SP. In complex events with numerous cosmic-ray and beam background particles, the simulated reconstruction and identification efficiency for triggered test-beam particles is above 80% for the majority of particle type and beam momentum combinations. Specifically, simulated 1 GeV/ c charged pions and protons are correctly reconstructed and identified with efficiencies of 86.1 $$\pm 0.6$$ ± 0.6 % and 84.1 $$\pm 0.6$$ ± 0.6 %, respectively. The efficiencies measured for test-beam data are shown to be within 5% of those predicted by the simulation. 
    more » « less
  5. ; ; ; ; ; ; (, Applied and Environmental Microbiology)
    Tortosa, Pablo (Ed.)
    ABSTRACT Bacteria shape interactions between hosts and fungal pathogens. In some cases, bacteria associated with fungi are essential for pathogen virulence. In other systems, host-associated microbiomes confer resistance against fungal pathogens. We studied an aphid-specific entomopathogenic fungus calledPandora neoaphidisin the context of both host and pathogen microbiomes. Aphids host several species of heritable bacteria, some of which confer resistance againstPandora. We first found that spores that emerged from aphids that harbored protective bacteria were less virulent against subsequent hosts and did not grow on plate media. We then used 16S amplicon sequencing to study the bacterial microbiome of fungal mycelia and spores during plate culturing and host infection. We found that the bacterial community is remarkably stable in culture despite dramatic changes in pathogen virulence. Last, we used an experimentally transformed symbiont of aphids to show thatPandoracan acquire host-associated bacteria during infection. Our results uncover new roles for bacteria in the dynamics of aphid-pathogen interactions and illustrate the importance of the broader microbiological context in studies of fungal pathogenesis. IMPORTANCEEntomopathogenic fungi play important roles in the population dynamics of many insect species. Understanding the factors shaping entomopathogen virulence is critical for agricultural management and for the use of fungi in pest biocontrol. We show that heritable bacteria in aphids, which confer protection to their hosts against fungal entomopathogens, influence virulence against subsequent hosts. Aphids reproduce asexually and are typically surrounded by genetically identical offspring, and thus these effects likely shape the dynamics of fungal disease in aphid populations. Furthermore, fungal entomopathogens are known to rapidly lose virulence in lab culture, complicating their laboratory use. We show that this phenomenon is not driven by changes in the associated bacterial microbiome. These results contribute to our broader understanding of the aphid model system and shed light on the biology of the Entomophthorales—an important but understudied group of fungi. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email