NCSI

   

XSEDE EMPOWER Positions


Shodor > NCSI > XSEDE EMPOWER > 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 / StatusSummaryCreator / Institution
We have recently introduced a scalable hybrid parallel algorithm, called phyNGSC, which allows fast compression as well as the decompression of big FASTQ datasets using distributed and shared memory programming models via MPI and OpenMP. This project will present the design and implementation of a novel parallel data structure which lessens the dependency on decompression and facilitates the handling of DNA sequences in their compressed state using fine-grained decompression in a technique that is identified as in compresso data processing. Our proposed structure and methodology will facilitate the enrichment of compressive genomics and sublinear analysis of big NGS datasets.
Sandino Vargas-Perez

Kalamazoo College
Kalamazoo, MI
We investigate various graphene defects as supports for single metal atoms. Previous investigations have been on Pt, as well as earth abundant metals like Fe, V, Ta, and Mo. We would like to investigate if the single atom stability we have found so far (especially for V and Ta) extends to a greater range of N-doped defect types.
Chloe Groome

University of California-Irvine
Irvine, CA
Students will be working on building a library of cache-oblivious data structures and measuring the performance under different workloads. We will first implement serial versions of the algorithms, and then implement the parallel version of several known cache oblivious data structures and algorithms. (We will be building off of our previous work from a previously funded XSEDE proposal -- "Parallel Cache Oblivious Data Structures and Algorithms") In our last session students have learned the foundations of performance engineering. I have two sets of students working on two projects. Group 1 (Ryan and Rob): - Students have read materials regarding Cache Oblivious Algorithms - We have additionally been reading the Open Data Structures book to gain further insights on data structure tradeoffs. - Students have spent a significant amount of time brushing up on C/C++ coding. - Students have been using the LLVM compiler to write optimization passes (what we are starting) in order to analyze programs. - We have had one publication accepted at a workshop Group 2 (Faridat): - Student has been practicing using super computing resources here at Northeastern (Our Discovery cluster) - Student has been sharpening both C++ and Python skills. - Student has implemented an edge detector algorithm for running with CUDA - Student then used profiling tools (based off XSEDE Readings from previous semester) and learned how to use NVProf to measure performance.
Michael Shah

Northeastern University
Boston, MA
Student(s) will be working on building tools to monitor performance sensitive applications. We will be building infrastructure to support writing 'performance assertions' that monitor the performance of high performance applications (games, graphics, volume visualizations, etc.) running with C/C++ code. Typically these applications are also highly parallelized (using frameworks like OpenMP or CILK), and can be difficult to track using regular profiling tools. Thus, we will be implementing a tool framework for monitoring the performance of these applications using the LLVM compiler framework.
Michael Shah

Northeastern University
Boston, MA
Develop large-scale analysis tools for molecular dynamics simulations. Perform molecular dynamics simulations of HIV-1 relevant systems. Collaborate with biophysical scientist in the design of experiments for the validation of the discoveries.
Juan Perilla

University of Delaware
Newark, DE
This project will conduct Micro-CT image segmentation and visualizations enhanced by machine learning techniques. A key challenge in using the X-ray CT imaging technique is to accurately segment material phases in contact so as to faithfully visualize the morphology, which provides the basis for image-based finite element modeling and analysis.
Jun Li

University of Massachusetts-Dartmouth
North Dartmouth, MA
The suicide rate in the US increased by 30% in the past decade. It has become a leading cause of death. Predicting suicide or suicide attempts are difficult. Community-based preemptive interventions have the potential to help targeted groups of high-risk people. This approach relies on the accurate characterization of suicide clusters and the timely monitoring of suicide clusters. We have developed some preliminary models with promising results.
Xinlian Liu

Hood College
Frederick, MD
Current approaches to tissue engineering fail because there is no central nervous system to orchestrate cell behavior in artificial cultures. We have created a microfluidics device capable of cell and fluid manipulations within living tissues. This project will develop a Model Predictive Controller (MPC) that will use a Recurrent Neural Network (RNN) in order to learn from empirical data and modulate the cell action based on real-time non-destructive feedback from 3D Lattice Light Sheet Microscopy (a Nobel Laureate technology available in the PI's lab) and chemical assaying. The ultimate goal is to achieve the ability to culture reproducible tissue patterns, specificed by the user prior to the at the beginning of the experiment. This technology will benefit patients in need of life-saving organ and tissue transplants.
Roman Voronov

New Jersey Institute of Technology
Newark, NJ
Turbulence is a complex multi-scale phenomenon characterized by chaos and randomness. However, turbulent flows are the rule, not the exception. Thus, a deep understanding of the transport phenomena driven by turbulence would permit the development of innovative flow/heat control tools for drag reduction, heat transfer enhancement, aeroacoustic noise control and mixing enhancement. In the last few decades, the discipline of thermal-fluid sciences has been reliant to High-Performance Computing (HPC) as a means of predicting flow behavior and understanding the thin zone around a solid immersed in a viscous fluid flow, the so called “boundary layer”. Moreover, turbulent boundary layers that evolve along the flow direction are ubiquitous. Computationally speaking, this type of boundary layer (i.e., spatially-developing boundary layer) poses an enormous challenge, due to the need for accurate and time dependent in flow turbulence information. Direct Numerical Simulation (DNS) is a numerical tool that resolves all turbulence scales; thus, it aims to provide high spatial/temporal flow data. In this project, we will utilize a large dataset of direct simulations of turbulent boundary layers in order to compute high order statistics of velocity, pressure and temperature fluctuations, such as two-point correlations and power spectra. The major objectives of the proposed study are three-fold: (i) to develop an efficient GPU-based C++ code as a postprocessing tool, (ii) to understand the effect of compressibility on the boundary layer structure, (iii) to evaluate the analogy between the velocity and thermal field.
Guillermo Araya

University of Puerto Rico-Mayaguez
Mayaguez, PR
This is follow-up research of Cognitive Sensing for Robot-human Interaction. Human-machine (Robot) interaction gains more and more interest recently thanks to the HPC. With speech recognition, face recognition and facial expression recognition, robots can “hear” and “feel” human beings. In this project, we further research on more than one robots that work cooperatively to finish one or more tasks such as reconstructing a 3D human skeleton model. The integration of data from different networked robots will be tested in XSEDE HPC.
Jiang Lu

University of Houston-Clear Lake
Houston, TX
This project will conduct finite element (FE) simulations to predict coupled electro-mechanical responses of carbon nanotube (CNT) reinforced polymer composites for small to large strain sensing applications. Realistic representative volume element (RVE) based on measurable microstructure information will be constructed, and the response will be simulated by concurrent electrical and mechanical FE analyses with the goal of exploring a wide range of strain sensing capabilities.
Jun Li

University of Massachusetts-Dartmouth
North Dartmouth, MA
This undergraduate student experience will involve developing parallel numerical codes for the simulation of energy transport and conversion in nanostructured semiconductor materials and devices. We will then proceed to study nanostructured thermoelectric devices for energy conversion with the goal of improving energy conversion efficiency and energy performance in nanoelectronics.
Zlatan Aksamija

University of Massachusetts-Amherst
Amherst, MA
Recommender systems provide an essential means of recommending items to users that meet their needs or preferences, which have proven to be beneficial in e-commerce society. However, an important but often overlooked information in recommender systems is a customer's portrait images. Recent progress in Computer Vision techniques and Deep Learning research has drastically improved the ability of a computer to sense and analyze data from a single 2D image. In this project, we will study a new knowledge-based recommender system using a single portrait image. In particular, we plan to use deep facial 2D/3D attributes to build a knowledge graph, and explore visualization results to examine physical correctness and visual plausibility of the generated recommendations.
Hao Ji

California State Polytechnic University-Pomona
Pomona, CA
Copper hyponitrites are proposed as intermediates in NO reduction, however mechanisms are not well understood. This apprentice, will use density functional theory to propose electronic structures for a series of copper hyponitrites that have been crystallographically characterized.
S. Chantal E. Stieber

California State Polytechnic University-Pomona
Pomona, CA
The undergraduate student will work with the mentor to learn how to use the high performance computing system for the code improvement and the visualization of large data set obtained from the compressible multi fluid numerical simulations.
Tulin Kaman

University of Arkansas
Fayetteville, AR

Not Logged In. Login