Welcome to the website for the Center for Ultra-scale Computing and Information Security at Northwestern University. The Center participants include faculty and researchers from the Department of Electrical Engineering and Computer Science, the Feinberg School of Medicine, and the Kellogg School of Management. CUCIS also has participants from various national labs such as Argonne National Laboratory (ANL), Sandia National Laboratory (SNL), Los Alamos National Laboratory (LANL), Lawrence Livermore National Laboratory (LLNL), and includes support from companies such as IBM, Intel, and Sun.

Today's "connect anytime and anywhere" society based on the use of digital technologies is fueling tremendous data growth and transforming the way our business, science, and digital technology-based world functions. Data in the terabytes range are not uncommon today and are expected to reach petabytes for many application domains in science, engineering, business, bioinformatics, and medicine in the near future. Ultra-scale refers to the two to three orders of magnitude increase in data sizes, computing power, and complexity of data compared to what is considered today. As data becomes so pervasive, assurance, trust and security of information becomes of paramount importance. Secure computing, therefore, must be considered as an important parameter in any research dealing with ultra-scale computing.

The goal of the Center for Ultra-scale Computing and Information Security (CUCIS) is to conduct highly innovative research in many synergistic areas of ultra-scale computing and information technologies. Furthermore, the goal of the center is to foster and enable inter-disciplinary research in computing technologies that scale to these levels. The CUCIS is directed by Prof. Alok Choudhary, and is presently supported by the NSF, DOE, Sandia National Labs, NASA, and Intel. To learn more, please use the navigation bar at the top of the page, or go directly to descriptions of our various research projects, a complete list of our publications, or more information about the group members.

Software Developed by the Center:

• Machine Learning for Materials
  • ElemNet -- Deep Learning the Chemistry of Materials From Only Elemental Composition and Enhancing Enhancing Materials Property Prediction by Leveraging Computational and Experimental Data using Deep Transfer Learning.
  • IRNet -- A General Purpose Deep Residual Regression Framework for Materials Discovery.
  • PADNet-XRD -- Peak Area Detection Network for Directly Learning Phase Regions from Raw X-ray Diffraction Patterns.
  • EBSD-indexing -- Extracting Grain Orientations from EBSD Patterns of Polycrystalline Materials Using Convolutional Neural Networks.
  • OPV_extratrees -- Machine Learning Framework for Estimating Efficiency of Organic Solar Cells using Extreme Random Forests
  • microstructure-optimization-ml -- Microstructure optimization of constrained design objectives using machine learning based feedback aware data generation
  • OPVPredictor -- Predictor for Highest Occupied Molecular Orbital (HOMO) Energy for Organic Photovoltaic Donor Molecules
  • SINet -- A Transfer Learning Framework for Organic Solar Cell Prediction using Multi Input Single Output Neural Networks
  • ml-iter-additive -- An iterative machine learning framework for predicting temperature profiles for an additive manufacturing process
  • CheMixNet -- Mixed DNN Architectures for Predicting Properties using Multiple Molecular Representations
  • microstructure-optimization-sampling -- Data Sampling Schemes for Galfenol Microstructure Optimization
  • Deep_learning_homogenization -- A Deep Learning Model For Homogenization of Two-Phase High-Contrast Three-dimensional Materials
• Machine Learning for Healthcare
  • ColonCancerOutcomeCalculator -- Colon Cancer Conditional Survival Outcome Calculator
  • SIGRNN -- Synthetic Minority Instances Generation in Imbalanced Datasets Using a Recurrent Neural Network Approach
• Parallel I/O
  • Parallel netCDF -- a parallel I/O library for accessing netCDF files
  • HDF5 Log-layout based VOL - an HDF5 VOL Plugin for storing datasets in a log-based layout.
  • Parallel HDF5 Dataset Concatenation for High Energy Physics Data Analysis
• Data Mining
  • NU-MineBench -- a collection of data mining algorithms and applications in sequential and parallel
  • Parallel Data Clustering Algorithms -- a collection of massively parallel data clustering algorithms, including DBSCAN, OPTICS, and PINK
  • Fast Max Clique Finder -- a software suite to find the maximum cliques of large sparse graphs. See a use case in social network data.
  • Parallel MEP -- a parallel community detection algorithm for graph data. The algorithm is based on the idea of maximizing equilibrium and purity of communities. The source codes are implemented in C and MPI to run on distributed-memory computer clusters.
  • PrunedSearch -- A machine learning based meta-heuristic approach for constrained continuous optimization.
• More can be found in https://github.com/NU-CUCIS.

Graduate Student Research Assistantship:

We are looking for graduate students who are interested in pursuing a Ph.D. degree and study the following research areas:

• High-performance data mining. The research topics include design and development of algorithm in the field of data mining/ machine learning in general. The students will have opportunity to work with application domain scientists in materials, climatology, cosmology, medicine/bioinformatics and others.

EECS Home | McCormick Home | Northwestern Home | Calendar: Plan-It Purple © 2011 Robert R. McCormick School of Engineering and Applied Science, Northwestern University "Tech": 2145 Sheridan Rd, Tech L359, Evanston IL 60208-3118  |  Phone: (847) 491-5410  |  Fax: (847) 491-4455 "Ford": 2133 Sheridan Rd, Ford Building, Rm 3-320, Evanston, IL 60208  |  Fax: (847) 491-5258 Email Director Last Updated: