XSEDE EMPOWER Positions

Please note that XSEDE EMPOWER ended with the conclusion of the XSEDE project on August 31, 2022. The following information is provided for archival purposes only.

This page provides a listing of positions that are under review for the next iteration of the program, as well as those that have been approved for current and previous iterations of the program. As part of the review and approval process, accepted student applicants are associated with an approved position.

If you are intending to mentor a student, you should create a position. If you also have a student in mind with whom you want to work, that student should submit an application. If you intend to mentor multiple students, please create a separate position for each student.

Project Title / Status	Summary	Creator / Institution
Sequential Monte Carlo (SMC) Methods for Data Assimilation Completed	The student will need to work on implementing sequential monte carlo (SMC) methods using XSEDE resources for data assimilation purposes. SMC methods will be implemented for simulating real-time applications (building occupancy estimation).	Sanish Rai Longwood University Farmville, VA
Creation of an Updated GalaxSee Interface Completed	GalaxSee is a program designed to teach issues in high performance computing surrounding the gravitational N-Body problem, and has been built in the past to run on Macs, PCs, Web Servers, in Java applets, on HPC clusters, and on the LittleFe parallel computing environment. We would like to update the current PC/Mac interface to run on the most recent OS versions, adding in the feature set parallelism that can be enabled for both multi-core through OpenMP and many-core if a suitable GPGPU is present. Additionally, we would like to add a steered mode to communicate with GalaxSee-HPC running on a remote XSEDE cluster, and update GalaxSee-HPC to make use of OpenMP and GPGPU based parallelism.	David Joiner Kean University Union, NJ
High-end Visualization of Turbulent Events in Wall-bounded Flows Completed	Turbulent boundary layers are mainly characterized by high Reynolds numbers, randomness, chaos and disparate range of turbulent length/time scales. Despite its chaotic behavior, investigation performed during the last six decades has conclusively demonstrated the presence of organized motions in turbulent boundary layers, so called coherent structures (CS). A coherent structure may be defined as a region or parcel of fluid where any fluctuating component of the flow is highly correlated with itself. These structures can be considered the building-blocks of turbulent boundary layers, and significant attention has recently been given to explanation of their creation, development and destruction. Therefore, this project involves time-dependent 3D scientific visualization of coherent structures and turbulent events from a Direct Numerical Simulation (DNS) database of spatially-developing turbulent boundary layers subject to streamwise favorable pressure gradient and crossflow jet by means of flow animation videos and a fully immersive approach or virtual reality (VR). The main purpose of the EMPOWER internship project will be to develop a post-processing code in C++ and Unity platforms with GPU capabilities in order to perform flow visualization in a virtual reality environment. In addition, Paraview and Blender software will be used for the creation of a series of videos to dynamically visualize turbulent events such as; a) streaky structure dynamics, b) streak breakups, c) turbulent bursts, d) symbiosis of streaks and streamwise vortices, and e) hairpin vortex dynamics. This is a continuation Spring-2018 EMPOWER internship project.	Guillermo Araya University of Puerto Rico-Mayaguez Mayaguez, PR
Accurate Hair Segmentation and Classification Using Deep Learning Completed	Many real-life applications such as human identification and beautification, demand accurate hair segmentation and classification in images. In this project, we plan to leverage state-of-the-art deep learning techniques to build accurate models for hair segmentation and classification. We will build deep learning models using TensorFlow, train them on XSEDE Bridges GPU clusters, and benchmark their performance with application cases running in real time using webcams and mobile devices.	Hao Ji California State Polytechnic University-Pomona Pomona, CA
Large-Scale Fashion Recommendation Systems Using TensorFlow and Spark Completed	Recommender systems provide an essential means of recommending products to users that meet their needs or preferences, which have proven to be beneficial in e-commerce society. An important research problem in modern recommender systems is how to integrate discriminative content features into the recommendation process. In this project, we plan to: 1) investigate effective and automatic deep learning models using TensorFlow to identify fashion items in street fashion images; and 2) design large-scale fashion recommender systems in Spark to accommodate the use of fashion features learned by deep learning models.	Hao Ji California State Polytechnic University-Pomona Pomona, CA
Parallel Implementation of Block Matrix Operations in Distributed Computing Systems Completed	This is a continuation of the previous XSEDE EMPOWER project "Novel Randomized Algorithms for Large-Scale Matrix Completion" conducted in summer 2018. In this project, we will develop, analyze, and implement block matrix computations using GraphX for distributed Spark systems, as well as their CUDA correspondents for distributed multiple-GPU systems. We will focus on a set of core matrix operations including sparse matrix multiplication, dense matrix multiplication, and Singular Value Decomposition, and evaluate their computational efficiency in many applications such as large-scale matrix completion and block Krylov subspace problems.	Hao Ji California State Polytechnic University-Pomona Pomona, CA
Parallel Implementation of Block Matrix Operations in Distributed Computing Systems Completed	This is a continuation of the previous XSEDE EMPOWER project "Novel Randomized Algorithms for Large-Scale Matrix Completion" conducted in summer 2018. In this project, we will develop, analyze, and implement block matrix computations using GraphX for distributed Spark systems, as well as their CUDA correspondents for distributed multiple-GPU systems. We will focus on a set of core matrix operations including sparse matrix multiplication, dense matrix multiplication, and Singular Value Decomposition, and evaluate their computational efficiency in many applications such as large-scale matrix completion and block Krylov subspace problems.	Hao Ji California State Polytechnic University-Pomona Pomona, CA