Nicholas A. Nystrom, Ph.D.

Curriculum Vitae

nystrom9095@gmail.com  ·  https://www.linkedin.com/in/nick-nystrom/

Expertise

Professional Experience

2019‑2020 Chief Scientist, Pittsburgh Supercomputing Center, Carnegie Mellon University
  • Developed a complete HPC+AI+Data ecosystem supporting research in medicine, science, and engineering, in coordination with diverse federal sponsors (NSF, NIH, DoD) and partnerships between academia and industry.
  • Architect, Principal Investigator, and Project Director for the Bridges-2 national supercomputer: $10M acquisition, with expected total funding of $22.5M (NSF Office of Advanced Cyberinfrastructure (OAC) award 1928147). Led development of the successful proposal for Bridges-2, working with vendor partners and the national scientific community, and directed the Bridges-2 acquisition and early scientific outreach. Developed in partnership with HPE, Bridges-2 is designed for rapidly evolving research, featuring 488 dual-socket AMD EPYC 7742 (Rome) nodes with 256 GB of RAM, 16 similar nodes with 512 GB of RAM, 4 large-memory nodes with 4 TB of RAM and 4 Xeon Platinum 8260M (Cascade Lake) CPUs, and 24 GPU nodes each with 8 NVIDIA Tesla V100-32GB SXM2 GPUs and 384–768 GB of CPU RAM. Bridges-2 is interconnected by Mellanox HDR-200 Infiniband, with dual rails interconnecting the GPU nodes to improve scaling for deep learning. Bridges-AI, which still has considerable value, is planned to be federated with Bridges-2 when Bridges is decommissioned. Bridges-2 introduces a flash filesystem to support training on large datasets and a hierarchical disk and tape storage system managed by HPE DMF (Data Management Framework).
  • Architect, Co-Principal Investigator, and Associate Director of Scientific and Broader Impact for Neocortex, a revolutionary AI supercomputer: $5M acquisition, with expected total funding of $11.25M (NSF Office of Advanced Cyberinfrastructure award 2005597). Neocortex integrates two Cerebras CS-1 systems, each with a Wafer Scale Engine (WSE) deep learning processor, with a large-memory (24TB) HPE Superdome Flex HPC server, with closely balanced bandwidths and integration with Bridges-2 for management of large data and complementary, general-purpose computing. This unique system is designed to enable research into scaling across multiple CS-1 systems, streaming data at high bandwidth from the SDFlex's large memory to each CS-1 independently or both together. It is intended to accelerate deep learning by a factor of up to 1,000 while maintaining familiar, easy-to-use interfaces (TensorFlow and PyTorch), to be followed by an SDK and API for researchers developing fundamental algorithms.
  • Director, Advancing Cancer Biology at the Frontiers of Machine Learning and Mechanistic Modeling, supported by the National Cancer Institute. NCI specifically invited me to direct this Innovation Lab, which brought together interdisciplinary experts who otherwise seldom work together. The Lab was extremely successful: Of the nine ideas generated for pilot projects, six were deemed worthy of funding. I issued four subawards through PSC for pilot projects that are expected to yield substantial R01 proposals, and NCI contributed additional support for two more.
  • Principal Investigator for Amplifying the Value of HuBMAP Data Through Data Interoperability and Collaboration, an NIH Common Fund Data Ecosystem (CFDE) project to apply machine learning to HuBMAP and Kids First (a complementary NIH CFDE consortium) data to understand causes of childhood cancers and structural birth defects and to amplify the FAIRness of HuBMAP data.
  • (Progression from prior appointment) Hardware/Software Architect and Principal Investigator for Human BioMolecular Atlas Program (HuBMAP) Infrastructure and Engagement, being developed with a hybrid HPC/cloud (currently AWS) approach to cost-effectively maximize capability, interoperability, and reliability: $5,710,740 awarded to date through the NIH Common Fund. The HuBMAP Consortium HuBMAP consists of 19 lead institutions and approximately 35 collaborating institutions, with 11 more lead institutions soon to be added. Led software development using an agile methodology to accommodate evolving requirements and the many developers of the Consortium. Led the core Infrastructure and Engagement Component to the inaugural HuBMAP data release (September 1, 2020). Complete information on HuBMAP is available through the HuBMAP Consortium Website, and HuBMAP data is available through the HuBMAP Data Portal.
  • (Progression from prior appointment) Architect, Principal Investigator, and Project Director for Bridges: directed production operations include systems support, scientific an educational outreach, and partnerships; successfully led annual NSF reviews of operations, and planned the transition to Bridges-2, including the federation of Bridges-AI with Bridges-2. Bridges has served over 2,100 research projects conducted by over 20,000 users at over 800 institutions (spanning academia, national labs and federal reserve banks, and industry) across the U.S. and their international collaborators.
  • Co-Principal Investigator for OCCAM+P38: Instantiating and Sustaining a Repository of Executable and Interactive Computer Architecture Experiments, supported by DoD. Directed the use of Bridges to support collaboration on computer architecture for Project 38, a set of architectural explorations involving DoD, the DOE Office of Science, and the DOE National Nuclear Security Administration (NNSA).
  • Strategic Planning, Center for AI Innovations in Medical Imaging (CAIIMI). CAIIMI is a new research center launched in January 2020 with members representing UPMC (University of Pittsburgh Medical Center hospital system), University of Pittsburgh, Carnegie Mellon University, and the Pittsburgh Supercomputing Center (myself). Worked with the CAIIMI director and team to plan an ambitious proposal and provided discretionary access to Bridges-AI in support of CAIIMI research.
  • Enabled urgent COVID-19 research. Provided discretionary allocations and priority scheduling on Bridges for the COVID-19 High Performance Computing Consortium (announced by the White House on March 23, 2020) as well as for COVID-19 projects external to the Consortium. The projects were in AI, genomics, and molecular dynamics. Significant advances made through those allocations included development of a fast (30-minute), low-cost, reliable test for COVID-19 (Mason group; Weill Cornell Medicine); a database of molecules that are candidates for therapeutics (Isayev group; Carnegie Mellon University); and another database of candidate therapeutics submitted to the Europe's JEDI Grand Challenge: Billion Molecules against COVID-19.
2017‑2019 Interim Director, Pittsburgh Supercomputing Center, Carnegie Mellon University
  • Led the PSC team of approximately 60 FTEs and PSC's interactions with its parent universities and external stakeholders.
  • Created a new Artificial Intelligence and Big Data group to amplify PSC's ability to generate new opportunities and make valuable contributions. Creation of the AI&BD group made possible Bridges-AI expansion of Bridges, the award for the highly innovative Neocortex system, and numerous research collaborations.
  • Co-led the successful proposal for Bridges-AI, a $1.8M expansion of Bridges that expanded aggregate AI capacity across NSF's national cyberinfrastructure by 283%. Bridges-AI introduced the NVIDIA DGX-2 and NVIDIA Tesla V100 (“Volta”) GPUs to the NSF research community to enable research using scalable AI.
  • Co-founded the Compass Consortium, an innovative program through which industry participants could evaluate emerging AI technologies for their specific domains, learn of best practices, and engage PSC and academic experts in Consortium (multi-partner, shared) and Pilot (single-partner) research projects.
  • Architect and Principal Investigator for Human BioMolecular Atlas Program (HuBMAP) Infrastructure and Engagement, being developed with a hybrid HPC/cloud approach to cost-effectively maximize capability, interoperability, and reliability: led development of the successful proposal to NIH, hybrid cloud hardware and interoperating microservices-based software architecture with federated Globus identity management, and the core Infrastructure and Engagement Component of the Consortium (~8 FTEs).
  • Principal Investigator for Challenges and Opportunities in Scientific Data Discovery and Reuse in the Data Revolution: Harnessing the Power of AI. Authored a successful proposal (NSF award 1839014, $50,000) to host the AIDR 2019: Artificial Intelligence for Data Discovery and Reuse conference, which was held on May 13–15, 2019 at Carnegie Mellon University.
  • Created the Pittsburgh Research Computing Initiative to catalyze data-, AI-, and HPC-driven collaboration and research across Carnegie Mellon University, the University of Pittsburgh, and the UPMC (University of Pittsburgh Medical Center) hospital system. Over 130 research groups participated.
  • (Progression from prior appointment) Architect, Principal Investigator, and Project Director for Bridges: Directed production operations include systems support, scientific an educational outreach, and partnerships.
  • (Progression from prior appointment) Co-Investigator for Big Data for Better Health (BD4BH). Led data infrastructure architecture for applying machine learning to genomic and imaging data for breast and lung cancer research, supported by the Pennsylvania Department of Health. Results included methods for detecting anomalous gene expression, indexing to search expression data for expressed viruses, identification of tumor progression features, etc.
  • (Progression from prior appointment) Principal Investigator for the Data Exacell. Led a team of ~5 FTEs to successfully conclude pilot projects. Results included demonstration of PSC's filesystem technology that was then deployed in Bridges, distributed deployment of the Galaxy framework that was transitioned to Bridges large-memory nodes for production use, and independent acquisition of the PghBio filesystem to serve The Cancer Genome Atlas and other biomedical data for Big Data for Better Health, the Center for Causal Discovery, the Center for AI Innovations in Medical Imaging, and other strategic projects.
  • (Progression from prior appointment) Co-Investigator and Software Performance Architect for the Center for Causal Discovery (CCD), an NIH Big Data to Knowledge (BD2K) Center of Excellence. Successfully concluded the NIH BD2K project, which produced numerous advances in algorithms for causal analysis of genomic, imaging, and time series data; improved understanding of signaling pathways and mutations; biomarkers of typical and atypical brain activity; and other areas.
  • Member of the Pittsburgh Quantum Institute (PQI): Advised on strategic directions for quantum applications and simulation.
2016‑2017 Sr. Director of Research, Pittsburgh Supercomputing Center, Carnegie Mellon University
  • Principal Investigator for the Data Exacell. Transitioned from Co-PI to PI. Led a team of approximately 5 FTEs to conduct pilot projects including the Pittsburgh Genome Research Repository (PGRR; thorough which a local, high-performance copy of The Cancer Genome Atlas (TCGA) was made available to researchers), a framework for reproducible science focusing on genomics (Galaxy), and a data-intensive workflow for radio astronomy (in collaboration with the National Radio Astronomy Observatory (NRAO)).
  • (Progression from prior appointment) Architect, Principal Investigator, and Project Director for the Bridges national supercomputer: directed installation and acceptance testing, including the world's first deployment of the Intel Omni-Path Architecture (OPA) fabric, and directed production operations, annual NSF reviews of the acquisition and operations, and approximately 30 PSC staff supported by the Bridges project.
  • (Progression from prior appointment) Co-Investigator for Big Data for Better Health (BD4BH). Led data and software architecture for applying machine learning to genomic and imaging data for breast and lung cancer research, supported by the Pennsylvania Department of Health.
  • (Progression from prior appointment) Co-Investigator and software performance architect for the Center for Causal Discovery (CCD), an NIH Big Data to Knowledge (BD2K) Center of Excellence. Advised on high-performance implementation of causal discovery algorithms, particularly for analysis of fMRI data to understand the brain causome. Directed implementation on Bridges of the Causal Web, a browser-based portal to democratize access to sophisticated algorithms (e.g. FGES) on large-memory nodes, effectively delivering HPC Software as a Service.
2015‑2016 Director of Research, Pittsburgh Supercomputing Center, Carnegie Mellon University
  • Architect, Principal Investigator, and Project Director for the Bridges national supercomputer, which pioneered the convergence of HPC, AI, and Big Data. Wrote the successful proposal that resulted in $20.9M of funding from the NSF Office of Advanced Cyberinfrastructure (award 1445606). The Bridges architecture was designed to make HPC and AI accessible to “nontraditional” communities and applications that had never used HPC before. It has been emulated multiple times around the world, and its success is substantiated by the recent NSF award for Bridges-2.
  • Co-Investigator and software performance architect for the Center for Causal Discovery (CCD), an NIH Big Data to Knowledge (BD2K) Center of Excellence. The CCD developed highly efficient causal discovery algorithms that can be practically applied to very large biomedical datasets and applied them to three distinct biomedical questions: cancer driver mutations, lung fibrosis, and the brain causome, as a vehicle for algorithm development and optimization. The CCD also disseminated causal discovery algorithms, software, and tools and trained scientists and biomedical investigators in the use of CCD tools. The CCD's graph algorithms made extensive use of PSC's large-memory Blacklight and Bridges systems, for which I was co-PI and PI, respectively. September 29, 2014–August 31, 2019.
  • Co-Investigator for Big Data for Better Health (BD4BH). Co-authored a successful proposal to the Pennsylvania Department of Health to apply state-of-the-art machine learning algorithms to genomic and imaging data for breast and lung cancer research, with Bridges providing data and AI capability to an interdisciplinary team including the University of Pittsburgh Department, CMU, UPMC, and PSC. Awarded for June 1, 2015–May 31, 2018.
  • Co-Principal Investigator for Open Compass. Co-authored the successful proposal “Open Compass: Leveraging the Compass AI Engineering Testbed to Accelerate Open Research” (NSF award 1833317; $300,000 awarded through NSF's EAGER program for high-risk, high-reward research; May 1, 2018–April 30, 2021), a project to evaluate emerging AI technologies for deep learning networks relevant to research (GCNs, 3D CNNs, CNNs for very large images, LSTMs for time series, etc.), develop and disseminate best practices, and conduct training.
  • Co-Principal Investigator for the Data Exacell. Authored the successful proposal “CIF21 DIBBS: The Data Exacell” (NSF award 1261721, $8,914,035, October 1, 2018–September 30, 2018; “DIBBs” = NSF's Data Infrastructure Building Blocks program). Originally envisaged as an exascale data management system to enable data-intensive research, the emphasis was shifted through the cooperative agreement to explore the potential of high-performance data analytics (HPDA) and novel data storage technologies. Initial computational technologies included large memory (Blacklight; 2×16TB cache-coherent shared memory), Sherlock (a multithreaded graph processor; YarcData uRiKa, with a primarily RDF+SPARQL user interface), and a SLASH2-based filesystem. I also authored a successful supplemental proposal to acquire PSC's first NVIDIA Tesla GPUs to add strong support for deep learning.
  • Co-Principal Investigator for Sherlock, a system for high-performance graph analytics. Authored the successful proposal to NSF (award 1234749, $1,226,000 September 1, 2012–August 31, 2015) to acquire and make available to the research community Sherlock, a Cray YarcData uRiKA data appliance consisting of a next-generation Cray XMT supercomputer (NG-XMT) running the uRiKA application architecture and augmented by Cray XT5 compute nodes to broaden the range of relevant applications to address the challenges of graph-based data analytics. It was based on the Cray XT5 infrastructure, specialized for graph analytics through the inclusion of Cray-proprietary Threadstorm 4.0 processors and AMD HyperTransport-attached SeaStar2 interconnect chips to provide a flat, globally-addressable memory.
  • Co-Principal Investigator for Blacklight, the world's largest shared-memory system. Authored the successful proposal “Very Large Shared Memory System for Science and Engineering” (NSF award 1041726, $2,942,517, August 1, 2010–July 31, 2014). Blacklight consisted of two SGI UV1000 systems, each with the maximum of 16TB (32TB aggregate) of hardware cache-coherent shared memory implemented on NUMAlink. It prioritized user productivity by allowing OpenMP, Java, and MATLAB applications to scale to 16TB and 2,048 cores. It also introduced the Nehalem microarchitecture via Intel Xeon X7560 CPUs to the national research community. Blacklight was the preferred system nationwide for large-scale genome sequence assembly and was also in high demand for data analytics. Its architecture, applications, and users informed the design of Bridges.
  • Co-Principal Investigator for SDCI HPC Improvement: High-Productivity Performance Engineering (Tools, Methods, Training) for NSF HPC Applications. Applied, evaluated, and hardened performance engineering tools including TAU, PAPI, PerfSuite, KOJAK, Scalasca, Vampir, and ipm for scientific applications with diverse execution profiles: AMR hydro+Nbody cosmology (ENZO), molecular dynamics (NAMD), and quantum simulation of nanoscale semiconductor devices (NEMO3D). Co-led tutorials in performance engineering at the SC08, SC09, LCI 2009, and LCI 2010 conferences.
  • Principal Investigator for EAGER: Exploring the Potential of “Native Client” for Computational Science. This NSF EAGER (rapid-turnaround, high-risk, high-reward) project tested the usability and affordability of Google's “Exacycle” cloud service using the “Native Client” (NaCl) programming model for scientific applications.
  • Lead, Advanced Computational Environments, DoD High Performance Computing Modernization Program (HPCMP) User Productivity Enhancement, Technology Transfer, and Training (PETTT). Worked with the HPTi/DRC/Engility (successive acquisitions) team and multiple universities to drive effective use of HPC resources by DoD research scientists.
  • (Progression from prior appointment) Led applications support for BigBen, PSC's Cray XT3 MPP system. Oversaw optimization of job placement in the Cray XT3's 3D torus, improving performance by 4.7–11.7%. Oversaw collaboration with a Cray hardware architect to optimize Seastar router settings, improving throughput for communications-intensive applications by up to 36% through age-based arbitration and changing the clock tick and 40% more by enabling and balancing across four virtual channels.
2004‑2020 Visiting Research Physicist, Physics Department, Carnegie Mellon University
  • Taught Advanced Computational Physics (MPI, high-performance 3D FFTs, data analytics, etc.).
  • Guest lectured at the CMU Physics Upper Class Colloquium.
  • Supported CMU Physics faculty in their use of PSC advanced computing resources and to compete for grants in Physics.
2004‑2005 Manager, Strategic Applications, Pittsburgh Supercomputing Center, Carnegie Mellon University
  • Led the Strategic Applications Group to identify and drive important areas of research, software development, and collaboration.
  • Led applications support for BigBen, PSC's Cray XT3 MPP system. When it was installed at PSC in 2005, BigBen was the world's first Cray XT3. Four full-scale applications — cosmology (GADGET), molecular dynamics (CHARMM), weather (WRF), and quantum chemistry (GAMESS) — were prepared in advance and demonstrated at the SC04 conference. Preparation began using an FPGA simulator for the Seastar router and the Catamount microkernel OS on machines at Cray. We collaborated with Sandia National Laboratory (SNL), where Cray Red Storm was developed, on scheduling, reliability, availability, and serviceability (RAS), and Catamount.
  • Led HCI study of programming language productivity, focusing on X10, UPC, and MPI for high concurrency applications, in collaboration with the IBM PERCS team and supported by the DARPA HPCS Program. Developed the SUMS methodology, which applies statistical techniques to comprehensive, fine-grained instrumentation of programmers’ activities and enables objective evaluation of soft ware development processes. In addition to precursor activities, we conducted a 4.5-day, IRB-approved human subjects study. Ours was the only university project invited to present at IBM s DARPA reviews.
  • (Progression from prior appointment) Co-Functional Area Point of Contact for Computational Chemistry and Materials Science, DoD High Performance Computing Modernization Program (HPCMP) User Productivity Enhancement and Technology Transfer (PET). Mentored an advanced intern in performance engineering and development of a GPU version of VASP (Vienna Ab initio Simulation Package, an important application for materials science and chemistry).
1998‑2003 Sr. Scientific Specialist, Pittsburgh Supercomputing Center, Carnegie Mellon University
  • Principal Investigator for Collaborative Research: ITR/AP: Novel Scalable Simulation Techniques for Chemistry, Materials Science and Biology. Co-developed a scalable, open source, ab initio molecular dynamics (Car-Parrinello MD) code with applications to chemistry, materials science and engineering, geoscience, and biology. Coded in Charm++, a parallel dialect of C++ that excels in latency hiding. NSF award 0121367, October 1, 2001–September 30, 2006, $206,773 (PSC portion).
  • Co-Functional Area Point of Contact for Computational Chemistry and Materials Science, DoD High Performance Computing Modernization Program (HPCMP) User Productivity Enhancement and Technology Transfer (PET). Led performance optimization, GPU acceleration, functionality enhancement, and applied research in collaboration with, and in support for, DoD research scientists. 2001–2009.
  • Ported and optimized a large number of applications and libraries for LeMieux, which introduced clusters of commodity CPUs (DEC Alpha EV68) to the NSF high performance computing community. LeMieux's 610 Compaq Alphaserver ES45 nodes were interconnected by dual-rail Quadrics. At the time of its installation, LeMieux was #2 on the Top500 list, following only LLNL's ASCI White system.
1994‑1998 Scientific Specialist, Pittsburgh Supercomputing Center, Carnegie Mellon University
  • Developed parallel versions of GAMESS (quantum chemistry), AMBER (molecular dynamics), and X-PLOR (X-ray crystallography) for the Cray T3D and T3E. Parallelization was implemented using PVM (prior to the development of MPI).
  • Taught and supported parallel programming for PSC's Cray T3D and T3E (Jaromir).
  • Provided tier-3 consulting in quantum chemistry and numerical methods.
  • Managed applications and libraries for PSC's Cray C916/512 supercomputer (Mario) and Cray T3D.
1992‑1994 Scientific Programmer, Pittsburgh Supercomputing Center, Carnegie Mellon University
  • Joint NSF-NASA Initiative in Evaluation (JNNIE): Evaluated the capabilities and deficiencies of scalable parallel computing architectures using a variety of application benchmarks.
  • Provided tier-3 consulting in quantum chemistry, numerical methods, and parallel programming.
  • Managed applications and libraries for PSC’s Cray C916/512 supercomputer (Mario) and Cray T3D.
1983‑1992 Software engineer, Self-employed
  • Software engineer: industrial simulation and metallurgical engineering, accounting systems, and enterprise databases.

Education

1985‑1992 Ph.D., Quantum Chemistry (also completed coursework in Physics)
University of Pittsburgh, Pittsburgh, PA
Advisor: Prof. Kenneth D. Jordan
1981‑1985 B.S., Chemistry, Math
University of Pittsburgh, Pittsburgh, PA
Advisor: Prof. Peter E. Siska

Financial Awards

I have attracted $98,539,209 in total funding for projects where I was Principal Investigator (PI) or Co-PI: $60,360,108 as PI and $38,179,101 as co-PI, from NSF, NIH, DOE, DoD, and other sources.

$1,000,000 NIH Common Fund (award in process as of 9/18/2020): Amplifying the Value of HuBMAP Data Through Data Interoperability and Collaboration. PI. September 1, 2020 – August 31, 2022.
$500,000 National Cancer Institute: Amplifying the Value of HuBMAP Data Through Data Interoperability and Collaboration. PI/Director. June 22, 2020 – June 21, 2021.
$22,500,000 NSF Office of Advanced Cyberinfrastructure (OAC) award 1928147: Category I: Bridges-2: Scalable Converged Computing, Data, and Analytics for Rapidly Evolving Science and Engineering Research. PI. October 1, 2019 – September 30, 2024 (estimated).
$11,250,000 NSF Office of Advanced Cyberinfrastructure (OAC) award 2005597: Category II: Unlocking Interactive AI Development for Rapidly Evolving Research. Co-PI. June 1, 2020 – May 31, 2025 (estimated).
$5,710,740 NIH OT2 (Common Fund) OD026675: Flexible Hybrid Cloud Infrastructure for Seamless Management of HuBMAP Resources. PI. $5,710,740 awarded as of 10/2020 (project years 1–3 of 4).
$300,000 NSF OAC-1833317: Open Compass: Leveraging the Compass AI Engineering Testbed to Accelerate Open Research. Co-PI. May 1, 2018 – April 30, 2020.
$50,000 NSF OAC-1839014: Challenges and Opportunities in Scientific Data Discovery and Reuse in the Data Revolution: Harnessing the Power of AI. PI. October 1, 2018 – September 30, 2019.
$20,895,167 NSF OAC-1445606: Bridges: From Communities and Data to Workflows and Insight. PI. December 1, 2014 – (estimated) November 30, 2020.
$8,914,035 NSF OAC-1445606: CIF21 DIBBS: The Data Exacell. PI. October 1, 2013 – September 30, 2019.
$35,923 NSF OAC-1419547: EAGER: Exploring the Potential of “Native Client” for Computational Science. PI. February 1, 2014 – January 31, 2016.
$5,042,791 Pennsylvania Department of Health: Big Data for Better Health (BD4BH) in Pennsylvania. Co-I. June 1, 2015 – May 31, 2018.
$11,000,000 NIH U54 HG008540: Center for Causal Modeling and Discovery of Biomedical Knowledge from Big Data (CCD). Co-I. February 1, 2014 – August 31, 2019.
$327,093 DOE National Energy Technology Laboratory (NETL): Enabling Multiphase Flow Research: Accelerating MFIX and Related Applications. PI. September 1, 2011 – August 31, 2013.
$219,377 DoD High Performance Computing Modernization Office User Productivity Enhancement, Technology, Transfer and Training (PETTT) Contract No. GS04T09DBC0017, Subcontract Agreement #HPTi-PETTT-CMU: Project #PP-CCM-KY02-125-P3: DoD HPC Software Development Environments Pilot; PETTT Scientific Python Toolset; PETTT Annual Technology Transfer Conference Support; VASP Performance/Scalability Improvement; HPC Courses Intro + MPI and IntroMP Hybrid Programming 2012. PI. September 1, 2009 – October 21, 2012.
$1,191,955 NSF 1234749: Enabling Productive, High-Performance Data Analytics. Co-PI. September 1, 2012 – August 31, 2015..
$2,942,517 NSF 1329280: A Very Large Shared Memory System for Science and Engineering. Co-PI. August 1, 2010 – July 31, 2014.
$2,218,065 NSF 0722072: SDCI HPC Improvement: High-Productivity Performance Engineering (Tools, Methods, Training) for NSF HPC Applications. Co-PI. November 1, 2007 – October 31, 2011.
$206,773 NSF CHE-0121273: Research: ITR/AP: Novel Scalable Simulation Techniques for Chemistry, Materials Science, and Biology. PI/Collaborative. (Total award: $4,233,773.) October 1, 2001 – September 30, 2006.
$500 University of Pittsburgh: Undergraduate Teaching Fellowship in Chemistry. August 1, 1984 – July 31, 1985.
$500 University of Pittsburgh: Undergraduate Teaching Fellowship in Chemistry. August 1, 1983 – July 31, 1984.

Awards and Honors

2019

Bridges-AI and Bridges contributed to two prestigious HPCwire Readers' Choice Awards, selected by an international nomination and voting process and presented at the the international SC18 conference:

  1. Best Use of HPC Life Sciences, Readers' Choice Award for development of deep learning techniques for precision immunotherapy, with Stony Brook Medicine.
  2. Top HPC-Enabled Scientific Achievement, Readers' Choice Award for supercomputer simulations to help reveal groundbreaking insight into gravitational waves by analyzing neutron star structures and mergers, with Perimeter Institute for Theoretical Physics (Ontario, Canada) and Theoretical Astrophysics Program (University of Arizona).
2018 Distinguished Alumnus, awarded by the University of Pittsburgh Department of Chemistry.
2018

Bridges contributed to a new record of 6 HPCwire awards at the international SC18 conference. HPCwire Readers' Choice Awards result from in international nomination and voting process.

  1. Best Use of AI, Readers’ Choice Award.
  2. Best Use of HPC in Life Sciences, Readers’ Choice Award.
  3. Best Use of HPC in Physical Sciences, Readers’ Choice Award.
  4. Best Use of High-Performance Data Analytics, Readers’ Choice Award.
  5. Best HPC Collaboration (Academia/Government/Industry),
    Readers’ Choice Award.
  6. Top HPC-Enabled Scientific Achievement, Editors’ Choice Award.
2017 Outstanding Leadership in HPC, HPCwire Readers' Choice Award, selected through an international competition organized by HPCwire and awarded at the international SC17 conference.

Bridges contributed to a record of 4 additional HPCwire awards, presented at the international SC17 conference:

  1. Best Use of AI, Readers' Choice Award.
  2. Best Use of HPC in Energy, Readers' Choice Award.
  3. Best Use of HPC in Life Sciences, Editors' Choice Award.
  4. Best Use of High-Performance Data Analytics, Editors' Choice Award.
1986 Inducted into the Phi Lambda Upsilon graduate chemistry honor society.
1984 Silverman Award in Chemistry.
1984 Inducted into the Omicron Delta Kappa honor society, recognizing achievement in scholarship and leadership.
1983 Kennedy Award in Mathematics.
1983 Inducted into the Druids honor society.
1982 Inducted into the Phi Eta Sigma honor society.
1981‑1985 National Merit Scholarship, University of Pittsburgh.

Publications

  1. P. A. Buitrago and N. A. Nystrom. In press. Neocortex and Bridges-2: A High Performance AI Ecosystem to Enable Research for Societal Good (paper accompanying invited keynote presentation). To appear in: H. Castro et al. (eds.) High Performance Computing: 7th Latin American Conference, CARLA 2020: Cuenca, Ecuador, September 2–4, 2020: Revised Selected Papers. Springer International Publishing.
  2. P. A. Buitrago and N. A. Nystrom. 2020. “Strengthening the Adoption of AI in Research and Cyberinfrastructure.” In: Pascucci, V., Altintas, I., Fortes, J., Foster, I., Gu, H., Hariri, S., Stanzione, D., Taufer, M., Zhao, X., Bremer, P. T., Moore, T., Nystrom, N. A., Petruzza, S., Ricart, G. (eds.) Report from the NSF Workshop on Smart Cyberinfrastructure 2020. Alexandria, Virginia (February 2020), http://smartci.sci.utah.edu/images/resources/NSFCI2020reportfinal3published.pdf.
  3. P. A. Buitrago, N. A. Nystrom, R. Gupta, R., and J. Saltz. 2020. Delivering Scalable Deep Learning to Research with Bridges-AI. In: Crespo-Mariño, J. L., Meneses-Rojas, E. (eds.) High Performance Computing: 6th Latin American Conference, CARLA 2019: Turrialba, Costa Rica, September 25–27, 2019: Revised Selected Papers. Communications in Computer and Information Science, vol. 1087, pp. 200–214. Springer International Publishing (2020). https://doi.org/10.1007/978-3-030-41005-6_14.
  4. M. P. Snyder, S. Lin, A. Posgai, M. Atkinson, A. Regev, J. Rood, O. Rozenblatt-Rosen, L. Gaffney, A. Hupalowska, R. Satija, N. Gehlenborg, J. Shendure, J. Laskin, P. Harbury, N. A. Nystrom, J. C. Silverstein, Z. Bar-Joseph, K. Zhang, K. Börner, Y. Lin, R. Conroy, D. Procaccini, A. L. Roy, A. Pillai, M. Brown, and Z. S. Galis. 2019. The human body at cellular resolution: the NIH Human Biomolecular Atlas Program. Nature, vol. 574, no. 7777, pp. 187–192, 2019. DOI: 10.1038/s41586-019-1629-x.
  5. N. Nystrom, P. Buitrago, and P. Blood. 2019. Bridges: Converging HPC, AI, and Big Data for Innovative Applications. In Contemporary High Performance Computing: From Petascale Toward Exascale, Volume 3. Jeffrey S. Vetter, Ed., CRC Computational Science Series, Taylor & Francis, Boca Raton. https://www.taylorfrancis.com/books/e/9781351036863/chapters/10.1201/9781351036863-14.
  6. J. U. Espino, J. Ramsey, K. Bui, C. Wongchokprasitti, Z. Yuan, M. Silvis, M. Davis, J. V. Shunfenthal, N. Nystrom, A. Labrinidis, P. K. Chrysanthis, and G. Cooper. 2017. A Software Suite for Causal Modeling and Discovery. DMMI-2017 (4th Workshop on Data Mining for Medical Informatics: Causal Inference for Health Data Analytics), Washington, D. C., November 4, 2017.
  7. S. Lu, K. N. Lu, S.-Y. Cheng, B. Hu, X. Ma, N. Nystrom, and X. Lu. 2015. Identifying Driver Genomic Alterations in Cancers by Searching Minimum-Weight, Mutually Exclusive Sets. PLoS Comput Biol 11(8): e1004257. DOI: 10.1371/journal.pcbi.1004257.
  8. N. A. Nystrom, M. J. Levine, R. Z. Roskies, and J. R. Scott. 2015. Bridges: A Uniquely Flexible HPC Resource for New Communities and Data Analytics. In Proceedings of the 2015 Annual Conference on Extreme Science and Engineering Discovery Environment (St. Louis, MO, July 26-30, 2015). XSEDE15. ACM, New York, NY, USA. DOI: 10.1145/2792745.2792775.
  9. Open Cyber Ontology Working Group (C. Chandler, E. Dull, D. Holmes, R. Kartch, N. Nystrom, S. Reinhardt, R. Techentin). 2014. Proposed Flow Data Ontology for RDF. Available at http://opencyberontology.sourceforge.net/.
  10. N. Nystrom, J. Welling, P. Blood, and E. L. Goh, “Blacklight: Coherent Shared Memory for Enabling Science”, in Contemporary High Performance Computing: From Petascale Toward Exascale, Jeffrey S. Vetter, Ed., CRC Computational Science Series, Taylor & Francis, Boca Raton, 2013, pp. 431−450.
  11. W. Kramer, R. Roskies, S. Lathrop, N. Nystrom, and S. Sanielevici, “Proceedings of the 6th Annual Extreme Scaling Workshop”, July 15-16, 2012, Chicago, Illinois, ACM Digital Library, https://dl.acm.org/citation.cfm?id=2462077.
  12. N. Nystrom, “Proceedings of the Petascale Applications Symposium: Multilevel Parallelism and Locality-Aware Algorithms”, June, 2007, Pittsburgh, Pennsylvania.
  13. N. Nystrom and D. O’Neal, “Applications Performance on the Cray T3E and T3E-900”, Cray User Group Proceedings, May 5-9, 1997, San Jose, California.
  14. N. Nystrom, D. Weisser, and J. Urbanic, “The SUMS Methodology for Understanding Productivity: Validation Through a Case Study Applying X10, UPC, and MPI to SSCA#1”, in The Third Workshop on Productivity and Performance in High-End Computing (PPHEC-6), Austin, Texas, 2006.
  15. D. Weisser, N. Nystrom, C. Vizino, S. T. Brown, and J. Urbanic, “Optimizing Job Placement on the Cray XT3”, CUG 2006 proceedings, Lugano, Switzerland, May, 2006.
  16. N. Nystrom, D. Weisser, J. Lim, Y. Wang, S. T. Brown, R. Reddy, N. T. B. Stone, P. Woodward, D. Porter, T. Di Matteo, L. V. Kalé, and G. Zheng, “Enabling Computational Science on the Cray XT3”, CUG 2006 proceedings, Lugano, Switzerland, May 2006.
  17. N. Nystrom, J. Urbanic, and C. Savinell*, “Understanding Productivity through Non-intrusive Instrumentation and Statistical Learning”, Proc. of the Productivity and Performance in High-End Computing (P-PHEC) workshop at the 11th International Symposium for High Performance Computer Architecture, San Francisco, February 13, 2005.
  18. H. N. Najm, J. H. Chen, J. F. Grcar, R. C. Armstrong, C. A. Kennedy, J. Ray, W. S. Koegler, A. E. Lutz, M. D. Allendorf, D. Klinke, A. H. McDaniel, N. Nystrom, R. Subramanya, “MPP Direct Numerical Simulation of Diesel Autoignition “,Sandia National Laboratory SAND2001-8075, Nov 2000.
  19. F. C. Wimberly, M. Lambert, N. Nystrom, A. Ropelewski, and W. Young, “Porting Third-Party Applications to the Cray T3D”, Parallel Computing 22, 1073-1089, 1996.
  20. N. Nystrom, W. S. Young, and F. C. Wimberly, “Methodologies for Developing Scientific Applications on the Cray T3D”, in Debugging and Performance Tuning for Parallel Computing Systems, M. L. Simmons, A. H. Hayes, J. S. Brown, and D. A. Reed, eds., IEEE Computer Society Press, Los Alamitos, California, 1996, pp. 279-297.
  21. W. Pfeiffer, S. Hotovy, N. A. Nystrom, D. Rudy, T. Sterling, M. Straka, “JNNIE: The Joint NSF-NASA Initiative on Evaluation,” San Diego Supercomputer Center Technical Report GA-A22123, 1995.
  22. C. Gonzalez, A. Ropelewski, S. Chitre, M. Lambert, N. Nystrom, D. Deerfield II, H. Nicholas Jr., and F. Wimberly, “Use of the Cray T3D Emulator in Implementing Scientific Applications”, CUG Spring ’93 Proceedings, Cray User Group, Montreaux, Switzerland, 1993.
  23. N. Nystrom, “Through-Bond and Through-Space Interactions in a Novel Class of Nonconjugated Polyenes,” Ph.D. Thesis, University of Pittsburgh, 1992.
  24. N. Nystrom, “Accurate Electronic Structure Calculations of Transition Metal Hydrides and Oxides by Relativistic Effective Potential Quantum Monte Carlo,” Ph.D. Proposal, University of Pittsburgh, 1990.

Presentations

Invited Presentations, Including Keynote and Plenary Presentations (Representative)

  1. Accelerating AI for Data-Driven Scientific Discovery, invited keynote at CARLA 2020 (Latin America High Performance Computing Conference), Cuenca, Ecuador, September 3, 2020.
  2. Balancing Extremes: New Directions in Expanding AI-Driven Research, invited presentation at the 3rd HPC Summer School: Bio & Data Science, Universidad de los Andes, Bogota, Colombia, July 28, 2020.
  3. Balancing Extremes: New Directions in Expanding AI-Driven Research, invited presentation for Stanford ME 344S: HPC-AI Summer Seminar Series, Stanford University (virtual), July 16, 2020.
  4. HuBMAP Infrastructure for Sharing Massive Biological Datasets, invited presentation at Intelligent Systems for Molecular Biology (ISMB) 2020, Montreal (virtual), July 14, 2020.
  5. A Journey in High Performance AI: Evolving Cyberinfrastructure, Democratizing Data, and Scaling AI to Catalyze Breakthroughs in Research, invited presentation for CMU ECE 18-847G: Special Topics in Computer Systems: Computational Problem Solving for Engineers, Carnegie Mellon University (virtual), April 28, 2020.
  6. A Journey in High Performance AI: Evolving Cyberinfrastructure, Democratizing Data, and Scaling AI to Catalyze Breakthroughs in Research, invited keynote for the HPC-AI Advisory Council Stanford Conference 2020, Sanford University (virtual), April 22, 2020.
  7. Multidisciplinary AI Today and Tomorrow with Bridges-AI and Bridges-2, invited presentation at NVIDIA GTC DC 2019, Washington, D.C., December 6, 2019.
  8. Multidisciplinary AI Today and Tomorrow with Bridges-AI and Bridges-2, invited presentation at the SC19 NVIDIA Theater, Denver, Colorado, November 20, 2019.
  9. Multidisciplinary AI Today and Tomorrow with Bridges-AI and Bridges-2, invited presentation at the SC19 HPE Theater, Denver, Colorado, November 19, 2019.
  10. Bridges-2: An Overview, invited presentation at the CASC Fall 2019 Membership Meeting, Alexandria, Virginia, September 26, 2019 (content presented by S. Sanielevici).
  11. New Systems, New Projects: Innovations in HPC, HPAI, and Data, invited presentation at the Bettis Atomic Power Laboratory Technical Briefing, Pittsburgh Supercomputing Center, September 19, 2019.
  12. Introducing Bridges-2, invited plenary at PEARC19, Chicago, Illinois, August 1, 2019.
  13. Bridges-2: A First Look, in Redefining Today's HPC, invited keynote at PEARC19 (invited to co-present by and with Trish Damkroger, VP, Intel Data Center Group), Chicago, July 31, 2019.
  14. Pioneering and Democratizing Scalable HPC+AI, invited keynote at the HPC-AI Advisory Council Stanford Conference 2019, Stanford University, February 15, 2019.
  15. The Future of Advanced Computing: Technology, Data, and Algorithms, invited presentation at the Council on Competitiveness BUILD 4 Advanced Computing Initiative, Pittsburgh, November 7, 2018.
  16. Advanced Computing, AI, and Data Solutions at PSC, invited seminar at the National Robotics Engineering Center (NREC), Carnegie Mellon University, September 5, 2018.
  17. Unleashing AI and Data for Medicine, invited keynote at the McGowan Institute for Regenerative Medicine 2018 Scientific Retreat, University of Pittsburgh, March 6, 2018.
  18. Converging High-Performance Computing, Artificial Intelligence, and Big Data, invited presentation at the Bayer IT Executive Leadership Team Meeting, Pittsburgh, March 5, 2018.
  19. Converging Life Sciences with Machine Learning and HPC, invited presentation at the LifeX Launch, University of Pittsburgh, December 11, 2017.
  20. AI Breakthroughs and Initiatives at the Pittsburgh Supercomputing Center, invited presentation at the 29th HPC User Forum, co-presented with Paola Buitrago, Milwaukee, Wisconsin, September 7, 2017.
  21. Bridges in Production – Experiences with an Architecture Optimized for Life Science, invited keynote at HP-CAST, Denver, November 10, 2017.
  22. Enabling Data-Driven Research and Communities, invited presentation at the Economic Research in High Performance Computing Environments Workshop, Center for the Advancement of Data and Research in Economics, Federal Reserve Bank of Kansas City, October 11, 2016, Kansas City, Missouri.
  23. A Converged HPC & Big Data Architecture in Production, invited keynote at HP-CAST, Salt Lake City, November 11, 2016.
  24. Bridges: Connecting Researchers, Big Data, and High-Performance Computing, invited lecture at the Café Scientifique, Carnegie Science Center, Pittsburgh, March 7, 2016.
  25. Bridges: A Uniquely Flexible HPC Resource for New Communities and Data Analytics, invited colloquium at the Computer Science Department, University of Pittsburgh, November 11, 2015.
  26. Big Data is Big News — and Big Careers!, invited lecture to inspire high school students at the Carnegie Science Center, Pittsburgh, November 5, 2015.
  27. A Uniquely Flexible HPC Resource for New Communities and Data Analytics, invited presentation at the Pittsburgh CIO Forum, Cranberry, Pennsylvania, September 15, 2015.
  28. Bridges, invited plenary panel presentation at XSEDE15, St. Louis, Missouri, July 29, 2015.
  29. A Uniquely Flexible HPC Resource for New Communities and Data Analytics, invited presentation at XSEDE15, St. Louis, Missouri, July 29, 2015.
  30. A New Data-Analytic Resource for the Digital Humanities, invited presentation at the Keystone Digital Humanities Conference, co-presented with Rick Costa, University of Pennsylvania, Philadelphia, July 22, 2015.
  31. Connecting Researchers, Data, and HPC, invited presentation at the PASSHE (PA State System of Higher Education) CAO (Chief Academic Officer) meeting, Indiana University, April 23, 2015.
  32. Bridges: An Introduction and The Data Exacell (DXC): Data Infrastructure Building Blocks for Coupling Analytics and Data Management, NITRD (Networking and Information Technology Research and Development) FASTER (Faster Administration of Science and Technology Education and Research) Briefing, National Science Foundation, Arlington, Virginia, December 15, 2014.
  33. The Data Exacell (DXC): Data Infrastructure Building Blocks for Coupling Analytics and Data Management, NITRD FASTER Briefing, NSF (Arlington, Virginia), December 15, 2014.
  34. Bridges: An Introduction and The Data Exacell (DXC): Data Infrastructure Building Blocks for Coupling Analytics and Data Management, University of Pittsburgh, October 15, 2014.
  35. Graph Analytics: An XSEDE Introduction, XSEDE ECSS Symposium, January 15, 2013.

Other Presentations (representative; see also Tutorials, below)

  1. High Performance AI: Democratizing Interdisciplinary Research, CMU 33-301 Upper Class Colloquium, Carnegie Mellon University, September 17, 2020.
  2. Introduction to Neocortex, Webinar, Virtual and YouTube, July 15, 2020.
  3. Bridges-2: Allocations Update, XRAC (XSEDE Resource Allocations Committee) Quarterly Meeting, Virtual, August 31, 2020.
  4. Bridges-2: Summary and Allocations, XRAC Reviewer Training, Virtual, August 7, 2020.
  5. Bridges-2: High Performance Data Analytics, AI, and Computing Available to HuBMAP, HuBMAP (Human BioMolecular Atlas Program) Consortium Sci/Tech Webinar Series, Virtual, July 13, 2020.
  6. Introducing Bridges-2, Webinar, Virtual and YouTube, May 28, 2020.
  7. Introducing Bridges-2, XSEDE Resource Allocation Committee (XRAC) Briefing, Chicago, Illinois, December 8, 2019 (content presented by R. Scibek).
  8. High Performance AI: A New Paradigm for Scientific Discovery, CMU Upper Class Colloquium, Carnegie Mellon University, September 12, 2019.
  9. Discovery at the Intersection of Science, Data, and Simulation, Junior/Senior Physics Colloquium, Carnegie Mellon University, October 25, 2018.

Interviews and Media Coverage (Representative)

  1. NPR: Scientists Use Pittsburgh Supercomputer To Create 3D Map Of The Human Body, September 9, 2020.
    → https://www.wesa.fm/post/scientists-use-pittsburgh-supercomputer-create-3d-map-human-body#stream/0
  2. InsideHPC: HPE to Build Bridges-2 Supercomputer at PSC, with Paola Buitrago, Alan George, and Manish Parashar, July 9, 2019.
    → https://insidehpc.com/2019/07/hpe-to-build-bridges-2-supercomputer-at-psc/?utm_source=feedburner
  3. Pittsburgh Post-Gazette: Personalized medicine, a human tissue atlas, and ‘prosperity to the region’; are all hopes for new $10M supercomputer, July 9, 2019.
    → https://www.post-gazette.com/business/healthcare-business/2019/07/09/National-Science-Foundation-10-million-grant-pittsburgh-supercomputing-center/stories/201907010096
  4. HPE Podcast: Inside story on HPC's role in the Bridges research project at Pittsburgh Supercomputing Center, with Paola Buitrago (PSC), March 14, 2018.
    → https://www.hpe.com/us/en/insights/articles/podcast-inside-story-on-hpcs-role-in-the-bridges-research-project-at-pittsburgh-supercomputing-center-1803.html
  5. BriefingsDirect: Inside story on HPC's role in the Bridges Research Project at Pittsburgh Supercomputing Center, with Paola Buitrago (PSC), November 21, 2017.
    → https://briefingsdirect.wordpress.com/2017/11/21/inside-story-on-hpcs-role-in-the-bridges-research-project-at-pittsburgh-supercomputing-center/.
  6. NPR: Pittsburgh Supercomputing Center Wins Awards For Poker Playing And DNA Sequencing Machines, November 14, 2017.
    → https://www.wesa.fm/post/pittsburgh-supercomputing-center-wins-awards-poker-playing-and-dna-sequencing-machines
  7. InsideHPC: AI Breakthroughs and Initiatives at the Pittsburgh Supercomputing Center, with Paola Buitrago (PSC), September 14, 2017.
    → https://insidehpc.com/2017/09/ai-breakthroughs-initiatives-pittsburgh-supercomputing-center/
  8. SiliconANGLE: AI supercomputing: From college labs to corporate offices, with Bill Mannel (HPE), June 9, 2017.
    → https://siliconangle.com/2017/06/09/ai-supercomputing-from-college-labs-to-corporate-offices-hpediscover/
  9. NPR: No Coding Skills Needed, Supercomputer Builds “Bridges” To Non-Traditional Users, January 24, 2017.
    → https://www.wesa.fm/post/no-coding-skills-needed-supercomputer-builds-bridges-non-traditional-users
  10. InsideHPC: Building Bridges to the Future, June 8, 2016.
    → https://insidehpc.com/2016/06/building-bridges-to-the-future/
  11. InsideHPC: PSC's Bridges Supercomputer Brings HPC to a New Class of Users, December 9, 2015.
    → https://insidehpc.com/2015/12/pscs-bridges-supercomputer-brings-hpc-to-a-new-class-of-users/
  12. InsideHPC: Hewlett Packard Enterprise, Intel and PSC: Driving Innovation in HPC, with Bill Mannel (HPE) and Charlie Wuischpard (Intel), November 17, 2015.
    → https://insidehpc.com/2015/12/hewlett-packard-enterprise-intel-and-psc-driving-innovation-in-hpc/
  13. HPCwire: A Conversation with PSC's Nick Nystrom, November 11, 2005.
    → https://www.hpcwire.com/2005/11/11/a_conversation_with_pscs_nick_nystrom-1/

Conference and User Group Leadership

  1. Program Committee, CARLA 2020 (Latin America High Performance Computing Conference), Cuenca, Ecuador, 2020.
  2. Program Committee, CARLA 2019 (Latin America High Performance Computing Conference), Turrialba, Costa Rica, 2019.
  3. Co-organizer, AMD HPC User Forum, 2020.
  4. Co-Program Chair, Artificial Intelligence for Data Discovery and Reuse (AIDR 2019), Carnegie Mellon University, Pittsburgh, 2019.
  5. Organizer, Omni-Path User Group, 2016.
  6. Co-organizer, Best Practices in Data Infrastructure Workshop, Pittsburgh Supercomputing Center, Pittsburgh, 2016.
  7. Co-organizer, XSEDE/Blue Waters Extreme Scaling Workshop: Heterogeneous Computing, Boulder, 2013.
  8. Co-organizer, XSEDE/Blue Waters Extreme Scaling Workshop: Algorithmic and Applications Challenges and Solutions in Large-scale Computing Systems with Limited Memory and I/O Bandwidth, Chicago, 2012.
  9. Co-organizer, TeraGrid/Blue Waters Symposium on Data-Intensive Analysis, Analytics, and Informatics, Pittsburgh, 2011.
  10. Logistics chair, PETTT Annual Technology Transfer Conference, Pittsburgh, 2011.
  11. Co-organizer, TeraGrid/Blue Waters Extreme Scaling Workshop: Parallel I/O, Austin, 2010.
  12. Co-organizer, TeraGrid/Blue Waters Extreme Scaling Workshop: Fault Tolerance and Resilience, Albuquerque, 2009.
  13. Co-organizer, TeraGrid/Blue Waters Extreme Scaling Workshop, Las Vegas, 2008.
  14. Chair, Petascale Applications Symposium: Multilevel Parallelism and Locality-Aware Algorithms, Pittsburgh Supercomputing Center, Pittsburgh, 2007.
  15. Co-organizer, TeraGrid/Blue Waters Extreme Scaling Workshop, 2007, Tucson.
  16. Program Committee, Third Workshop on Productivity and Performance in High-End Computing (PPHEC-06), Austin, 2006.
  17. Technical Chair, DOE Computer Graphics Forum, 2003, Lake Tahoe.

Teaching and Education

The Pittsburgh Supercomputing Center is currently not a degree-granting department, so creativity was required to provide opportunities in HPC and AI to students. I actively involved 23 undergraduate and graduate student interns in research projects that resulted in publications, a DOE fellowship, graduate school, and successful careers, and I have also served on PhD committees.

PhD Committees

  1. Ph.D. Committee Member for Evgeny Karataev, University of Pittsburgh, Ph.D. Information Science, 2016.
    Dissertation: Advanced distributed data integration infrastructure and research data management portal.
    First position after graduation: Senior Software Developer at Nielson.
  2. Ph.D. Committee Member for Wei Yu, Carnegie Mellon University, Ph.D. Electrical and Computer Engineering, 2011.
    Dissertation: Performance Portable Tracking of Evolving Surfaces.
    First position after graduation: Algorithmic Trading Researcher at DRW Trading Group, Chicago.

Graduate and Undergraduate Student Interns

  1. Co-Mentor, PSC AI Internship for Ya Ting (Tina) Chang, Carnegie Mellon University, MSIT: Business Intelligence and Data Analytics, 2018.
    • Project: Machine learning identification of small molecule drug candidates for targeted cancer therapeutics. Built machine learning models for small molecule drug candidates binding to micro-RNAs, achieving 87% accuracy, a 6% improvement over previously published results. A Pittsburgh biotech startup leveraged the results in funding pitches.
    • First position after graduation: Data Engineer, The Walt Disney Company.
  2. Co-Mentor, PSC AI Internship for Chia-Hua (Alice) Lee, Carnegie Mellon University, MSIT: Business Intelligence and Data Analytics, 2018.
    • Project: Dashboards for analyzing large-scale operational data on Bridges.
    • First position after graduation: Senior Data Scientist, Wayfair.
  3. Co-Mentor, PSC AI Internship for Anand Sakhare, Carnegie Mellon University, MSIT: Business Intelligence and Data Analytics, 2018.
    • Project: Scaling deep learning training with TensorFlow and Horovod.
    • First position after graduation: Big Data Architect, Amazon Web Serices (AWS).
  4. Co-Mentor, PSC AI Internship for Eric Su, Carnegie Mellon University, Undergraduate, B.S. Physics, 2018.
    • Project: Machine learning analysis of system logs to improve system performance and usability. Developed analytics to improve scheduling and user experience on HPC systems.
    • First position after graduation: Software Engineer, Roblox (San Jose).
  5. Mentor, PSC Bridges Internship for Ishtar Nyawĩra, University of Pittsburgh, 2016–2018.
    • Project: Deep Learning Reconstruction of the Connectome. Developed deep learning approaches to automating segmentation of high-resolution scanning electron microscope images of brain tissue, resulting in two peer-reviewed conference papers.
    • First position after graduation: Business System Developer, CUNA Mutual Group.
  6. Mentor, PSC Bridges Internship for Kristi Bushman, University of Pittsburgh (undergrad), 2017–2018.
    • Project: Deep Learning Reconstruction of the Connectome. Developed deep learning approaches to automating segmentation of high-resolution scanning electron microscope images of brain tissue, resulting in two peer-reviewed conference papers.
    • First position after graduation: Research Staff, Pitt Smart Living Project.
  7. Mentor, PSC Bridges Internship for Liyunshu (Iris) Qian, Carnegie Mellon University (undergrad), 2016–2017.
    • Project: Deep Learning Reconstruction of the Connectome. Developed deep learning approaches to automating segmentation of high-resolution scanning electron microscope images of brain tissue, resulting in a peer-reviewed conference paper.
    • First position after graduation: Admitted to CMU's aster of Science in Computational Finance program.
  8. Mentor, PSC Bridges Internship for Annie Zhang, Carnegie Mellon University (undergrad), 2016–2017.
    • Project: Deep Learning Reconstruction of the Connectome. Developed deep learning approaches to automating segmentation of high-resolution scanning electron microscope images of brain tissue, resulting in a peer-reviewed conference paper.
    • First position after graduation: Software Engineer, at Smarkets.
  9. Mentor, PSC Internship for Tisha Sethapakdi, Carnegie Mellon University (undergrad), 2014.
    • Project: Scaling Graph Analytics. Studied graph analytics using semantic web technologies (RDF and SPARQL) on Sherlock, PSD's unique YarcData Urika system, and compared performance of large-scale graph analytics on Sherlock and PSC's Blacklight system, which also supported graph algorithms through extremely large shared memory.
    • First position after graduation: Admitted to CMU's Master of Human-Computer Interaction program.
  10. Mentor, PSC Internship for Christopher Bradfield, University of North Carolina at Chapel Hill (undergrad), 2013.
    • Project: Scaling Graph Analytics. Applied Sherlock to the traveling salesman problem and cluster finding in astrophysics.
    • First position after graduation: Software Engineer, Square.
  11. Mentor, PSC Internship for Maxwell Hutchinson, Carnegie Mellon University (undergrad), 2008–2009.
    • Project: Performance Engineering and GPU Acceleration of VASP. Analyzed performance of parallel programming models (UPC, MPI). Then, in joint collaboration with Mike Widom group (CMU Physics) and supported by DoD PET, developed the GPU implementation of the electronic structure code VASP.
    • First positions after graduation: DOE fellowship at the University of Chicago (PhD Condensed Matter Physics), followed by Scientific Softare Engineer at Citrine Informatics.
  12. Mentor, PSC Internship for Jordan Soyke, University of Pittsburgh (undergrad), 2006–2007.
    • Project: HPC Application Performance Engineering.
    • First position after graduation: Software Engineer, Google.
  13. Mentor, PSC Internship for Jon Itturalde, Universidad del CEMA (undergrad), Argentina, 2008.
    • Project: Performance Analysis of Parallelization Techniques. Analyzed the performance of PGAS, POSIX threads (p-threads), and MPI implementations of a variety of algorithms on Blacklight, a large shared-memory system.
    • First position after graduation: Digitalization Manager, Irsa Propiedades Comerciales, Argentina.
  14. Mentor, PSC Internship for Uriel Jaroslawski, Universidad del CEMA (undergrad), Argentina, 2008.
    • Project: Performance Analysis of Parallelization Techniques. Analyzed the performance of PGAS, POSIX threads (p-threads), and MPI implementations of a variety of algorithms on Blacklight, a large shared-memory system.
    • First position after graduation: Senior Manager Product Development, Mercado Livre Brasil.
  15. Mentor, PSC Internship for Courtney Machi, University of Pittsburgh (undergrad), 2006.
    • Project: Productivity of HPCS Programming Languages. Implemented SUMS, a framework to study human productivity aspects of the X10 computer language relative to UPC and MPI. Supported by DARPA HPCS, through the IBM PERCS project.
    • First position after graduation: Director of Product Management, Toptal.
  16. Mentor, PSC Internship for Victor Puchkarev, University of Pittsburgh (undergrad), 2005–2006.
    • Project: Productivity of HPCS Programming Languages. Implemented SUMS, a framework to study human productivity aspects of the X10 computer language relative to UPC and MPI. Supported by DARPA HPCS, through the IBM PERCS project.
    • First position after graduation:
  17. Mentor, PSC Internship for William Clark, University of Pittsburgh (undergrad), 2004–2005.
    • Project: Productivity of HPCS Programming Languages. Implemented SUMS, a framework to study human productivity aspects of the X10 computer language relative to UPC and MPI. Supported by DARPA HPCS, through the IBM PERCS project.
    • First position after graduation:
  18. Mentor, PSC Internship for James Larkby-Lahet, University of Pittsburgh (undergrad), 2004–2005.
    • Project: Productivity of HPCS Programming Languages. Implemented SUMS, a framework to study human productivity aspects of the X10 computer language relative to UPC and MPI. Supported by DARPA HPCS, through the IBM PERCS project.
    • First position after graduation: Senior Software Engineer, Plantronics.
  19. Mentor, PSC Internship for Christina Savinell, University of Pittsburgh (undergrad), 2004.
    • Project: Productivity of HPCS Programming Languages. Implemented SUMS, a framework to study human productivity aspects of the X10 computer language relative to UPC and MPI. Supported by DARPA HPCS, through the IBM PERCS project.
    • First position after graduation: Multimedia Manager, Pittsburgh Symphony Orchestra.
  20. Mentor, PSC Internship for Austin McKinley, Carnegie Mellon University (undergrad), 2004.
    • Project: Productivity of HPCS Programming Languages. Implemented SUMS, a framework to study human productivity aspects of the X10 computer language relative to UPC and MPI. Supported by DARPA HPCS, through the IBM PERCS project.
    • First position after graduation: Software Engineer, Facebook.
  21. Mentor, PSC Internship for Nicholas Baldy, University of Pittsburgh (undergrad), 2004.
    • Project: Productivity of HPCS Programming Languages. Implemented SUMS, a framework to study human productivity aspects of the X10 computer language relative to UPC and MPI. Supported by DARPA HPCS, through the IBM PERCS project.
  22. Mentor, PSC Internship for Mark Smorul, University of Pittsburgh (undergrad), 2004.
    • Project: Productivity of HPCS Programming Languages. Implemented SUMS, a framework to study human productivity aspects of the X10 computer language relative to UPC and MPI. Supported by DARPA HPCS, through the IBM PERCS project.
    • First position after graduation: Project Software Engineer, Philips Respironics.
  23. Mentor, PSC Internship for Kiran Penumacha, Carnegie Mellon University (undergrad), 2003.
    • Project: HPC Software Administrator. Installed and tested software infrastructure.
    • First position after graduation: Co-founded Valence Energy Inc. (which was quickly acquired), then Founder & CEO of Grene Robotics, Hyderabad, India.

Lectures and Courses (See also: Other Presentations)

  1. Guest Lecturer: Special Topics in Computer Systems: Computing for Engineers, ECE 18847G, Carnegie Mellon University.
  2. Guest Lecturer: Big Data in Sustainability, IS 67-361, Carnegie Mellon University.
  3. Guest Lecturer: Advanced Computational Physics, PHY 33-456, Carnegie Mellon University.
  4. Mentor: Pittsburgh Supercomputing Center High School Initiative in Computational Science, Pittsburgh Supercomputing Center, 1993–1995.
  5. Teaching Assistant: Physical Chemistry, University of Pittsburgh, 1985–1992.
  6. Teaching Assistant: Computational Methods in Chemistry, University of Pittsburgh,
    1986–1987.
  7. Teaching Fellow: Undergraduate General Chemistry, 1983–1985.
  8. Mentor: Chemistry Olympics and College in High School high school programs, University of Pittsburgh.

Tutorials

  1. Using POINT Performance Tools: TAU, PerfSuite, PAPI, Scalasca, and Vampir (full day), S. Shende, D. Cronk, N. Nystrom, and R. Liu, 11th LCI International Conference on High-Performance Clustered Computing, Pittsburgh, PA, March 9, 2010.
  2. Productive Performance Engineering of Petascale Applications with POINT and VI-HPS (full day), SC09, Portland, Oregon, November 16, 2009.
  3. Parallel Performance Evaluation Tools: TAU, PerfSuite, PAPI, Scalasca (full day), S. Shende, A. Malony, R. Kufrin, R. Reddy, N. Nystrom, and S. Moore, 10th LCI International Conference on High-Performance Clustered Computing, Boulder Colorado, March 9, 2009.
  4. Productive Performance Engineering of Petascale Applications with POINT (full day), S. Shende, A. Malony, S. Moore, N. Nystrom and R. Kufrin, SC08, Austin, TX, November 17, 2008.

Advisory Boards

2020‑present Functional Interpretation of Alzheimer's Loci across Cell Types, Age, and DNA Damage Advisory Committee
Funded by the Pennsylvania Department of Health, this recent, valuable project in Alzheimer’s disease (AD) research is applying machine learning and high-throughput techniques to link AD-associated noncoding variants with endophenotypes, measure the impact of aging and DNA damage on cells, and identify downstream consequences associated with AD noncoding variants, as well as to provide training to underrepresented minority students.
2017‑present Hewlett Packard Enterprise User Advisory Board
Provide input on future application requirements and trends for HPC, AI, and data-intensive systems.
2017‑2020 National Center for Multiscale Modeling of Biological Systems (MMBioS) Executive Committee
The National Center for Multiscale Modeling of Biological Systems (MMBioS), an NIH Biomedical Technology and Research Resource (BTRR), develops tools to advance and facilitate cutting-edge research at the interface between high performance computing and the life sciences.
2016‑2018 DesignSafe-CI Advisory Board
The Natural Hazards Engineering Research Infrastructure (NHERI) is a distributed, multi-user, national facility that provides the natural hazards engineering community with state-of-the-art research infrastructure. DesignSafe embraces a cloud strategy for the big data generated in natural hazards engineering research and supports research workflows, data analysis, and visualization.
2009‑2010 Graph 500 Steering Committee
The Graph 500 benchmarks address algorithms that are important for high performance data analytics: concurrent search, optimization (single source shortest path), and edge-oriented (maximal independent set). These algorithms have different computational requirements from HPC systems that are designed for applications involving dense operations. Examples of application areas are cybersecurity, medical informatics, data enrichment, social networks, and symbolic networks.

Federal Review Panels

2009‑present Panel chair and reviewer for NSF proposals in advanced cyberinfrastructure and various scientific fields.

Professional Memberships

2010‑present American Association for the Advancement of Science (AAAS)
2002‑present Association for Computing Machinery (ACM)
2002‑present IEEE Computer Society
2002‑present Institute of Electrical and Electronics Engineers (IEEE)