BEGIN:VCALENDAR VERSION:2.0 PRODID:Linklings LLC BEGIN:VTIMEZONE TZID:America/New_York X-LIC-LOCATION:America/New_York BEGIN:DAYLIGHT TZOFFSETFROM:-0500 TZOFFSETTO:-0400 TZNAME:EDT DTSTART:19700308T020000 RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU END:DAYLIGHT BEGIN:STANDARD TZOFFSETFROM:-0400 TZOFFSETTO:-0500 TZNAME:EST DTSTART:19701101T020000 RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20210402T160558Z LOCATION:Poster Module DTSTART;TZID=America/New_York:20201118T083000 DTEND;TZID=America/New_York:20201118T170000 UID:submissions.supercomputing.org_SC20_sess340@linklings.com SUMMARY:ACM Student Research Competition Posters Display DESCRIPTION:ACM Student Research Competition: Graduate Poster, ACM Student Research Competition: Undergraduate Poster, Posters\n\nMassively Parallel Exact Histogram Equalization\n\nKirkpatrick\n\nHistogram specification is performed by transforming an image so that the image's histogram matches a target histogram. Exact methods of histogram specification result in les s loss of information but are orders of magnitudes slower than the classic al methods. Applications such as real-time medical i...\n\n--------------- ------\nRandomized Cholesky Factorization in Parallel\n\nLiang\n\nLarge sp arse SDD (symmetric diagonally dominant) linear systems arise from solving elliptic PDEs and graph problems. Exact Cholesky factorization leads to e xcessive fill-in, especially in 3D. Our approach is to subsample the (dens e) Schur complement update and construct an (approximate) sparse preco...\ n\n---------------------\nReducing Data Motion of Lattice Boltzmann Simula tions through Application of Boundary Conditions on GPUs\n\nDube\n\nLattic e Boltzmann simulations are commonly used for solving many computational f luid dynamics problems. The main computation of many of these simulations are separated into two parts: collision and streaming of the lattice Boltz mann method particle distribution function, and application of boundary .. .\n\n---------------------\nVisualizing Metagenomic Data in R Using Jetstr eam\n\nLeffler\n\nMetagenomes consist of the total genome content collecte d from an environmental sample containing bacterial, archaeal, and viral s equences present. These datasets are complex and can be overwhelming to vi sualize. Using multiple visualization methods benefits researchers by allo wing them to perform e...\n\n---------------------\nAutomatic Capture and Classification of Frog Calls\n\nForan, Underwood\n\nGlobal frog population s are threatened by an increasing number of environmental threats such as habitat loss, disease, and pollution. Traditionally, in-person acoustic su rveys of frogs have measured population loss and conservation outcomes amo ng these visually cryptic species. However, these method...\n\n----------- ----------\nFuture-Proof Your Research: Designing for Replicability and Re producibility\n\nBrunkan\n\nComputer Science is Agile. Iteration after ite ration, moving quickly to solve new problems, uncover new questions, find the next big thing. Hardware, software, libraries, datasets, experiments - technology becomes outdated almost as soon as it’s released. So why save code? Why share code? For replic...\n\n---------------------\nInvestigatin g Inconsistencies in Single-Cell RNA Sequencing Data\n\nSingh\n\nSingle-ce ll RNA sequencing (scRNA-seq) has emerged as a robust method for computati onal genomics. It combines cell biology with computer science in order to analyze thousands of cells at a single-cell level, allowing us to observe gene expression levels and how they differ across diverse samples. Th...\n \n---------------------\nNetGraf: A Collaborative Network Monitoring Stack for Network Experimental Testbeds\n\nKaur\n\nNetwork performance monitori ng collects heterogeneous data such as network flow data to give an overvi ew of network performance, and other metrics, necessary for diagnosing and optimizing service quality. However, due to disparate and heterogeneity, to obtain metrics and visualize entire data from s...\n\n----------------- ----\nHPC Data-Center Cooling Performance and Design\n\nBohn\n\nCooling pe rformance and design in a data-center are critical to its successful opera tion. There are many contributing factors to the efficiency and effectiven ess of the cooling in data centers, both in the design phase and in the po st-analysis to determine and improve performance. The visualization ...\n\ n---------------------\nHPC Rankings Based on Real Applications\n\nJarmusc h, Baker\n\nPerformance benchmarks are used to stress test hardware and so ftware of large scale computing systems. A corporation known as SPEC has d eveloped a benchmark suite, SPEC ACCEL, consisting of test codes represent ative of kernels in large applications. This project ranks the published r esults from ACCE...\n\n---------------------\nModeling Power Usage for the SZ Lossy Compressor on HPC Systems\n\nWilkins\n\nLossy compressors are be coming more prevalent due to the increasing volumes of data produced in HP C systems. Compressors are integral to computational workflows to assist i n data I/O: transporting smaller amounts of data is more time and energy e fficient. By modeling power consumption of lossy compr...\n\n------------- --------\nAI-Guided Adaptive Multiscale Modeling of Platelet Dynamics\n\nH an, Zhang, Deng\n\nWe developed an AI-guided adaptive multiple time steppi ng algorithm to model platelet activation, adhesion and aggregation, compl ex dynamical processes that cause physiological reactions including cardio vascular diseases and stroke. The dynamics spans 6 spatial and 9 temporal scales. Our algorithm c...\n\n---------------------\nUsing Machine Learnin g for OpenMP GPU Offloading in LLVM\n\nMishra\n\nOpenMP 5.0 provides featu res to exploit the compute power within the node of today's leadership cla ss facilities. Among these features, the GPU offloading directives are key to take advantage of heterogeneity on modern machines. These features pla ce the domain scientists with portability challenges,...\n\n-------------- -------\nMulti-Agent Meta Reinforcement Learning for Packet Routing in Dyn amic Network Environments\n\nSun\n\nTraffic optimization challenges, such as flow scheduling and completion time reducing, are difficult online deci sion-making problems in wide area networks. Previous works apply heuristic s that rely on full knowledge of the system to design optimization algorit hms. In this work, we explore building a...\n\n---------------------\nA Co mputer Vision and AI Based Solution to Determine the Change in Water Level in Stream\n\nChandra\n\nFlooding is one of the most dangerous weather eve nts today. Between 2015-2019 on average, it has caused more than 130 deat hs every year in the USA alone. World Health Organization has reported tha t, between 1998-2017, floods have affected more than 2 billion people worl dwide. The devastating nature...\n\n---------------------\nContainerized E nvironment for Reproducibility and Traceability of Scientific Workflows\n\ nOlaya García\n\nScientists rely on simulations to study natural phenomena . Trusting the simulation results is vital to develop sciences in any fiel d. One approach to build trust is to ensure the reproducibility and tracea bility of the simulations through the annotation of executions at the sys tem-level; by the gen...\n\n---------------------\nActiveness-Based Data R etention Recommender for HPC Facilities\n\nZhang\n\nThe storage system of many high-performance computing (HPC) facilities faces an increasingly cha llenging goal of meeting unlimited data growth with limited storage capaci ty growth. The data retention policy plays a vital role in addressing such challenge. However, most existing data retention polici...\n\n----------- ----------\nOceananigans.jl: Improving Climate Model Accuracy with Fast an d Friendly Geophysical Fluid Dynamics on GPUs\n\nRamadhan\n\nClimate model s cannot resolve every cloud in the atmosphere and every wave in the ocean , but their collective effects can be important for global climate. Climat e models rely on ad-hoc surrogate models to resolve the effects of these s ub-grid scale physical processes. Many existing surrogate models ...\n\n-- -------------------\nEnhancing IoT Anomaly Detection Performance for Feder ated Learning\n\nWeinger\n\nWhile federated learning has gained great atte ntion for mobile computing with the benefits of scalable cooperative learn ing and privacy protection capabilities, there still exist a great deal of technical challenges to make it practically deployable. For instance, dis tribution of the training proces...\n\n---------------------\nOptimizing P erformance in Power Bounded GPU Computing\n\nAllen\n\nPower is consistentl y considered as a top challenge in HPC. To address the power challenge, po wer efficient GPU accelerators have become a standard component in mainstr eam systems and contribute the majority of compute capacity. Nevertheless, GPUs are still constrained by limited permissible power a...\n\n--------- ------------\nSparsity-Aware Distributed Tensor Decomposition\n\nMiao\n\nT ensors are used by a wide range of applications as data structures to mode l multi-dimensional data. Tensor decomposition is a class of methods for latent data analytics. This work presents a sparsity-aware tensor decompos ition on a distributed memory system. We optimize the CANDECOMP/PARAFAC de com...\n\n\nTag: Student Program\n\nRegistration Category: Tech Program Re g Pass, Exhibits Reg Pass END:VEVENT END:VCALENDAR