Opportunities for Collaboration with IBM Allen Malony Dept
Opportunities for Collaboration with IBM Allen Malony Dept. of Computer and Information Science Neuroinformatics Center University of Oregon Eugene, Oregon, USA malony@cs. uoregon. edu
Outline Goals Areas of opportunity CIS undergraduate / graduate educational programs Lundquist College of Business Resource development Research areas Computational science TAU parallel performance system BBMI / UO Neuroinformatics Center Integrative Science Complex 2
Goals for IBM Collaboration Establish and foster a relationship with IBM Beaverton Human opportunities Enhance educational programs in CIS and UO Create and strengthen institutional partnerships Provide new internship opportunities to CIS students Improve funding support through fellowships Partner on broader education and diversity initiatives Research and development opportunities Explore strategic development and research alliances Link UO research activities with IBM efforts Work on joint projects and write joint proposals 3
CIS Undergraduate Program Major requirements consist of CIS core courses Completion of a traditional CIS or interdisciplinary track CIS core courses Lower division Fundamental CIS 210 -12, Intro to CS CIS 313, Data structures Math 231 -33, Discrete Math organization Upper division CIS 314, Computer CIS 315, Intro to algorithms CIS 415, Operating systems CIS 422, Software methodology WR 320, Scientific / Technical CIS 425, Programming languages WR 321, Business communication Writing (either) 4
CIS Undergraduate Tracks Traditional Foundations Software Development Computer Networks Database and Informatics Interdisciplinary Bioinformatics Computational Biology Computational Arts: Multimedia Business Information Technology 5
CIS Minor, CIT Minor, and MACS Major Mathematics and Computer Science (MACS) major Joint major with Math department Explore computer science with foundation in mathematics Develop team players for information-base occupations CIS Minor Introduces theories and techniques of computer science Develop skills applicable to major discipline Computer Information Technology (CIS) Minor Train students to work with evolving technologies Business databases, computer networks, web applications, software systems 6
CIS and Lundquist College of Business Information Technology track in the CIS major Approved by College of Arts and Science for Fall ‘ 06 CIS major (slightly modified) plus minor in business Will have capstone projects for the students in this track CIT minor Work with Faculty in Decision Sciences to update and revamp Add more classes that business students can take in CIS Computer Ethics (already OKed) plus two additional Certificate program in Business Information Technology Offered by CIS CIT minor + three additional 400 level classes 7
CIS and Business Interactions Leadership in Action practicum (Spring ‘ 06) Experimental teams (2 business majors + 2 CIS majors) Teams go out to do a project for local non-profit (Mt. Pisgah) Technology Entrepreneursip Program Technology transfer program with UO departments and PNNL Prof. Jun Li's m. SSL secure authentication protocol selected Symantec Favorable response to CIT/CIS business interdisiplinary student Has hired several students from UO/CIS Portland IT Association funds for IT scholarships Outstanding CIT minors First awards Fall ‘ 06 8
Human Collaboration Ideas Career mentoring Career Mentorship Colloquium (Winter ‘ 07) IBM guest speakers Internship program Create co-op / work-study program 6 month (fall/spring + summer) IBM intern 6 month (fall+winter / winter+spring) academic program Summer research internships IBM fellowships Extreme Blue Research fellowships (T. J. Watson, Almaden, …) IBM post-docs 9
Resource Initiatives Improve educational and research resources UO wide CIS specific Enhance CIS tracks Establish bridges from CIS curriculum to research Enhance interdisciplinary activities within UO Add value to existing initiatives Create opportunities for establishment of new projects 10
Digital Collaborative Work-Learn-Play Space Collaboration with AAA Fab Lab Modular, reconfigurable high-tech spaces State of the art digital hardware, software, tools, and toys Support Interdisciplinary cross-campus innovation Experimental classes Special lectures and events Hands-on experimentation with newest gadgets and gizmos High-tech social events 3 D movies, graphics film fest Digital relaxation and conversation Phase I conceptualizaiton and planning 11
CIS Scientific Innovation Center Support interdisciplinary computer science Informatics Computational science Resource development Phase 1 (complete) NSF MRI grant ($1 M) ICONIC Grid Phase II (underway) Visualization Lab ($100 K) 3 x 4 tiles 24” LCD display rear projection, stereo display Phase III (with sponsors) 12
UO ICONIC Grid 13
World Community Grid Invitation to be a partner in the WCG CIS and UO excited to join Involvement ideas Introduce to CIS students in undergraduate/graduate courses Feature participation on CIS webpage Allocation of dedicated servers Contribution of application ideas Help promote broader UO involvement NIC involvement Participation in human brain science projects Integration with ICONIC Grid 14
Computational Science Math Integration of computer science in traditional Biology scientific disciplines Computer Geoscience Increasingly accepted Neuroscience Science model of scientific research Application of high-performance Psychology computation, algorithms, networking, Paleontology database, and visualization Parallel and grid computing Integrated problem-solving environments Computer science research at the core Support for interdisciplinary science 15
Computational Science Projects at UO Geological science Model coupling for hydrology Bioinformatics Zebrafish Information Network (ZFIN) Evolution of gene families Oregon Bioinformatics Toolkit Neuroinformatics Paleontology Dinosaur skeleton and motion modeling Artificial intelligence and data mining Semantic web, data ontologies, and information integration Oregon Computational Science Institute 16
HPC Research Project Areas at UO Parallel modeling and simulation Computational sciences Bioinformatics Scientific problem solving environments Grid computing Parallel performance evaluation and tools Parallel language systems Tools for parallel system and software interaction Source code analysis Parallel component software Data mining 17
HPC Research Affiliations at UO Strong associations with DOE national laboratories Los Alamos National Lab, Lawrence Livermore National Lab, Sandia National Lab, Argonne National Lab, Pacific Northwest National Lab DOE funding Office of Science, Advance Scientific Computing Research Accelerated Strategic Computing (ASC/NNSA) NSF funding Academic Research Infrastructure Major Research Instrumentation 18
TAU Performance System Tuning and Analysis Utilities (14+ year project effort) Performance system framework for HPC systems Integrated, scalable, flexible, and parallel Open source toolkit for performance problem solving Instrumentation, measurement, analysis, and visualization All computing platforms, processor architectures, OSs IBM, HP, Sun, Linux, SGI, Cray, Apple, Windows, … Many programming languages and compilers C, C++, Fortran 90/95, UPC, HPF, Java, Open. MP, Python IBM, Intel, PGI, GNU, Sun, Microsoft, SGI, Cray, HP, … Partners: LLNL, ANL, Research Center Jülich, LANL 19
TAU Projects TAU performance system Para. Prof: parallel profile analysis and visualization Perf. DMF: performance data management Perf. Explorer: performance data mining KTAU: Linux kernel performance monitoring Integration with the Eclipse Platform Open Trace Format (OTF) for scalable parallel tracing Performance analysis of large-scale applications CCSM / ESMF / WRF climate/earth/weather simulation National laboratory applications DOE Sci. DAC and Do. D PET (Programming Environments and Tools) 20
TAU Eclipse Integration Eclipse GUI integration of existing TAU tools New Eclipse plug-in for code instrumentation Integration with CDT and FDT Java, C/C++, and Fortran projects Can be instrumented and run from within eclipse Each project can be given multiple build configurations corresponding to available TAU makefiles All TAU configuration options are available Paraprof tool can be launched automatically
TAU Eclipse Integration TAU experimentation TAU configuration 22
TAU Eclipse Future Work Development of the TAU Eclipse plugins ongoing Java and the CDT/FDT Planned features include: Full integration with the Eclipse Parallel Tools Project (PTP) Database storage of project performance data Refinement of the plugin settings interface easier selection of TAU runtime and compiletime options Accessibility of TAU configuration and commandline tools via the Eclipse UI 23
Zepto. OS and TAU DOE OS/RTS for Extreme Scale Scientific Computation OS research for petascale systems Zepto. OS project scalable, adaptive components for petascale architectures Argonne National Laboratory and University of Oregon Kernel-level performance monitoring OS component performance assessment and tuning KTAU (Kernel Tuning and Analysis Utilities) integration of TAU infrastructure in Linux kernel port to 32 -bit and 64 -bit Linux platforms include in Zepto. OS and port to IBM BG/L (I/O node) 24
Linux Kernel Profiling using TAU – Goals Fine-grained kernel-level performance measurement Parallel applications Support both profiling and tracing Both process-centric and system-wide view Merge user-space performance with kernel-space User-space: (TAU) profile/trace Kernel-space: (KTAU) profile/trace Detailed program-OS interaction data Including interrupts (IRQ) Analysis and visualization compatible with TAU 25
KTAU Architecture 26
KTAU Future Work Dynamic measurement control Enable/disable events w/o recompilation or reboot Add new performance data sources Look into hardware counters Improve user-space integration Full callpaths and phase-based profiling Merged user/kernel traces Integration with monitoring technology Super. Mon, MRNet, TAUg New porting efforts IA-64, PPC-64 and AMD Opteron System characterization studies 27
IBM Involvement TAU runs on all IBM platforms and processors p. Series, x. Series, BG/L (64 K system at LLNL) All national laboratory IBM systems run TAU Porting to Cell processor IBM Petascale Tools Workshop (‘ 05 and ‘ 06) IBM T. J. Watson Research Center (David Klepacki) Student participation (K. Huck, A. Nataraj) IBM HPCS project TAU integration in Eclipse Parallel Tools Platform (PTP) Open Software Development Lab (OSDL) University of Oregon membership 28
TAU Support Acknowledgements Department of Energy (DOE) Office of Science MICS, Argonne National Lab ASC/NNSA University of Utah ASC/NNSA Level 1 ASC/NNSA, Lawrence Livermore National Lab Department of Defense (Do. D) HPC Modernization Office (HPCMO) Programming Environment and Training (PET) NSF Software and Tools for High-End Computing Research Centre Juelich Los Alamos National Laboratory Para. Tools, Inc. 29
Neuroinformatics Center (NIC) UO Brain, Biology, and Machine Initiative (BBMI) Interdisciplinary research in cognitive neuroscience, biology, physics, and computer science Human neuroscience focus Computational science applied to human neuroscience Tools to help understand dynamic brain function Tools to help diagnosis brain-related disorders HPC simulation, large-scale data analysis, visualization Integration of neuroimaging methods and technology Interaction with Electrical Geodesics, Inc. Technology transfer to Cerebral Data Systems 30
NIC Projects Advanced statistical signal analysis (PCA, ICA) Hi. Per. SAT high-performance signal analysis toolkit parallelize ICA, wavelet, and other algorithms APECS automated protocol for EEG component separation Computational human head / brain modeling Conductivity modeling and source localization modeling Image segmentation and brain structure extraction Distributed and grid computing GEMINI Mc (Matlab concurrent) 31
Electrical Geodesics Inc. (EGI) EGI Geodesics Sensor Net Dense-array sensor technology 64/128/256 channels 256 -channel geodesics sensor net Ag. Cl plastic electrodes Carbon fiber leads Net Station Advanced EEG/ERP data analysis Stereotactic EEG sensor registration Research and medical services Epilepsy diagnosis, pre-surgical planning 32
Dipole Sources in the Cortex Scalp EEG is generated in the cortex Interested in dipole location, orientation, and magnitude Cortical sheet gives possible dipole locations Orientation is normal to cortical surface Need to capture convoluted geometry in 3 D mesh From segmented MRI/CT Linear superposition 33
Building Computational Brain Models MRI segmentation of brain tissues Conductivity model Measure head tissue conductivity Electrical impedance tomography small currents are injected between electrode pair resulting potential measured at remaining electrodes Finite element forward solution Source inverse modeling Explicit and implicit methods Bayesian methodology 34
GEMINI “Grid-based Environments and Methods for Integrated Neuroimaging” Dynamic neuroimaging algorithms and visualization Grid-based integration (processing and data sharing) High-end tool integration and environments Neuroinformatics data ontologies 35
Technology Transfer in Human Neuroscience UO’s BBMI is conducting pioneering research and development in human neuroscience, genetics and proteomics, and computational science for future neurological medicine and health care Greater precision and speed in brain imaging has high research and medical relevance Integrated medical imaging (EEG/MEG, MRI, radiology) Automatic image assessment (detection and diagnosis) Neurological evaluation and surgical planning Linking of genetics factors with complex cognitive traits (personality, learning, attention) has potential for therapies and pharmaceutical clinical drug development 36
Leveraging Internet, HPC, and Grid Computing Telemedicine imaging and neurology Distributed EEG and MRI measurement and analysis Neurological medical services Shared brain data repositories Remote and rural imaging capabilities Need to enhance HPC and grid infrastructure in Oregon Build on emerging web services and grid technology Establish HPC resources with high-bandwidth networks Create institutional and industry partnerships UO is working closely with EGI to develop high-end EEG analysis services framework Pilot neuroimaging services model on ICONIC Grid 37
Assistive Technology and Brain Injury Research Technology for people with cognitive impairments Navigation Email Trimet Multi-disciplinary research Prof. Steve Fickas, CIS Wearable Computing Lab Prof. Mc. Kay Sohlberg, Education NSF grants Cog. Link, Inc. Startup company http: //www. go-outside. org/ 38
Genomics and Bioinformatics Research in comparative genomics analyzes similarities and differences between orthologous genes ortholog = “same word” Zebrafish, salmon, and other teleost fish often have two orthologs of a single human gene UO project: software to scan human chromosomes, identify co-orthologs in zebrafish Studying co-orthologs improves John Postlethwait our ability to understand functions(Neuroscience) of genes, potential medical applications John Conery (CIS) Salmon calcitonin is up to 50 times more effective than human calcitonin in treating osteoporosis 39
Dinosaur 3 D modeling Dino. Morph modeling engine Paleontology-based Reconstructs true dimensions, poses, flexibility, and movements Dinosaur species Other domestic, wild, and fanciful animals Kaibridge, Inc. Startup company Dinosaur interactive museum exhibits Dinosaur educational software BBC online mystery game Photo by Rick Edwards, AMNH, 2006 Computational Paleontology 40
UO Integrative Science Discoveries of the 21 th century will arise from integrative science Evolution in how science is done UO CAS is positioned at the forefront Tradition of interdisciplinary and interdepartmental research Institutes and centers report to UO Vice President for Research Only Oregon member of Association of American Universities Carnegie Foundation rates UO as having high research activity Highlights Institute of Molecular Biology Institute of Theoretical Science Materials Science Institute of Neuroscience Oregon Center for Optics Institute of Cognitive & Decision Science Zebrafish Information Network Brain, Biology, and Machine Initiative Lewis Center for Neuroimaging / NIC ONAMI 41
UO Integrative Science Complex (ISC) Catalyst for innovation Bring leading researchers from diverse disciplines together Biology Neuroimaging Geological Science Cognitive Neuroscience Materials Science Computer Science Green Nanoscience Optics Genomics Molecular Biology Two phase plan Phase I: “From Nano to Neuro” Fall 2007, $16 M Phase II Estimated completion 2010 -11, $50 M 42
UO Integrated Science Complex - Phase I Signature research center (Lokey Laboratory) Oregon Nanoscience and Microtechnologies Institute (ONAMI) High-tech extension service for ONAMI “Best in class” facility for nanoscale research Gold standard for minimal vibration Encased in bedrock 19 feet below ground Open to all Oregon institutions Space for corporations to lease Proprietary or collaborative research Lorry Lokey Chairman and CEO, Business Wire 43
UO Integrated Science Complex - Phase II New ISC building Uniting Streisinger, Huestis, and Deschutes Halls New faculty research facilities New classroom space Facilitate interaction and integrations across scientific disciplines Phase II Above-ground entrance to the Lokey Laboratories Focus on biological sciences Phase I 44
- Slides: 44