Search

Your search keyword '"A. Houle"' showing total 346 results
Start Over Author "A. Houle" Publisher ieee
346 results on '"A. Houle"'

4. Imaging Performance of the Fully Assembled Ultra-High Resolution (UHR) Brain PET scanner

Author

Loignon-Houle, F., primary, Toussaint, M., additional, Beaudoin, J.-F., additional, Gaudreault, M., additional, Doyon, V., additional, Leroux, J.-D., additional, Auger, E., additional, Thibaudeau, C., additional, Arpin, L., additional, Croteau, E., additional, Espinosa-Bentancourt, E., additional, Samson, A., additional, Bouchard, J., additional, Espagnet, R., additional, Viscogliosi, N., additional, Pepin, C. M., additional, Labrecque, V., additional, Paulin, C., additional, Marin, T., additional, Ouyang, J., additional, Normandin, M., additional, Tétrault, M.-A., additional, Michaud, J.-B., additional, Fontaine, R., additional, El Fakhri, G., additional, and Lecomte, R., additional

Published
2023
Full Text
View/download PDF

5. Microliter in vivo Brain PET Imaging with the Ultra-High Resolution (UHR) Scanner

Author

Doyon, V., primary, Sarrhini, O., additional, Loignon-Houle, F., additional, Toussaint, M., additional, Auger, E., additional, Thibaudeau, C., additional, Croteau, E., additional, Lavallée, E., additional, Dumulon-Perreault, V., additional, Beaudoin, J.-F., additional, Leroux, J.-D., additional, Bouchard, J., additional, Samson, A., additional, Espagnet, R., additional, Fontaine, R., additional, and Lecomte, R., additional

Published
2023
Full Text
View/download PDF

7. Toward a second generation of Metascintillators using the Purcell effect

Author

Lecoq, P., primary, Kaminer, I., additional, Schultzman, A., additional, Schütz, R., additional, Kurman, Y., additional, Lahav, N., additional, Dosovitskiy, G., additional, Bekenstein, Y., additional, Konstantinou, G., additional, Latella, R., additional, Zhang, L., additional, Gonzalez, A. J., additional, Loignon-Houle, F., additional, and Benlloch, J. M., additional

Published
2023
Full Text
View/download PDF

8. Mapping of Internet “Coastlines” via Large Scale Anonymized Network Source Correlations

Author

Jananthan, Hayden, primary, Kepner, Jeremy, additional, Jones, Michael, additional, Arcand, William, additional, Bestor, David, additional, Bergeron, William, additional, Byun, Chansup, additional, Davis, Timothy, additional, Gadepally, Vijay, additional, Grant, Daniel, additional, Houle, Michael, additional, Hubbell, Matthew, additional, Klein, Anna, additional, Milechin, Lauren, additional, Morales, Guillermo, additional, Morris, Andrew, additional, Mullen, Julie, additional, Patel, Ritesh, additional, Pentl, Alex, additional, Pisharody, Sandeep, additional, Prout, Andrew, additional, Reuther, Albert, additional, Rosa, Antonio, additional, Samsi, Siddharth, additional, Trigg, Tyler, additional, Wachman, Gabriel, additional, Yee, Charles, additional, and Michaleas, Peter, additional

Published
2023
Full Text
View/download PDF

9. Deployment of Real-Time Network Traffic Analysis Using GraphBLAS Hypersparse Matrices and D4M Associative Arrays

Author

Jones, Michael, primary, Kepner, Jeremy, additional, Prout, Andrew, additional, Davis, Timothy, additional, Arcand, William, additional, Bestor, David, additional, Bergeron, William, additional, Byun, Chansup, additional, Gadepally, Vijay, additional, Houle, Micheal, additional, Hubbell, Matthew, additional, Jananthan, Hayden, additional, Klein, Anna, additional, Milechin, Lauren, additional, Morales, Guillermo, additional, Mullen, Julie, additional, Patel, Ritesh, additional, Pisharody, Sandeep, additional, Reuther, Albert, additional, Rosa, Antonio, additional, Samsi, Siddharth, additional, Yee, Charles, additional, and Michaleas, Peter, additional

Published
2023
Full Text
View/download PDF

10. Focusing and Calibration of Large Scale Network Sensors Using GraphBLAS Anonymized Hypersparse Matrices

Author

Kepner, Jeremy, primary, Jones, Michael, additional, Dykstra, Phil, additional, Byun, Chansup, additional, Davis, Timothy, additional, Jananthan, Hayden, additional, Arcand, William, additional, Bestor, David, additional, Bergeron, William, additional, Gadepally, Vijay, additional, Houle, Micheal, additional, Hubbell, Matthew, additional, Klein, Anna, additional, Milechin, Lauren, additional, Morales, Guillermo, additional, Mullen, Julie, additional, Patel, Ritesh, additional, Pentland, Alex, additional, Pisharody, Sandeep, additional, Prout, Andrew, additional, Reuther, Albert, additional, Rosa, Antonio, additional, Samsi, Siddharth, additional, Trigg, Tyler, additional, Yee, Charles, additional, and Michaleas, Peter, additional

Published
2023
Full Text
View/download PDF

11. pPython Performance Study

Author

Byun, Chansup, primary, Arcand, William, additional, Bestor, David, additional, Bergeron, Bill, additional, Gadepally, Vijay, additional, Houle, Michael, additional, Hubbell, Matthew, additional, Jananthan, Hayden, additional, Jones, Michael, additional, Klein, Anna, additional, Michaleas, Peter, additional, Milechin, Lauren, additional, Morales, Guillermo, additional, Mullen, Julie, additional, Prout, Andrew, additional, Reuther, Albert, additional, Rosa, Antonio, additional, Samsi, Siddharth, additional, Yee, Charles, additional, and Kepner, Jeremy, additional

Published
2023
Full Text
View/download PDF

16. GraphBLAS on the Edge: Anonymized High Performance Streaming of Network Traffic

Author

Jones, Michael, Kepner, Jeremy, Andersen, Daniel, Buluc, Aydin, Byun, Chansup, Claffy, K, Davis, Timothy, Arcand, William, Bernays, Jonathan, Bestor, David, Bergeron, William, Gadepally, Vijay, Houle, Micheal, Hubbell, Matthew, Jananthan, Hayden, Klein, Anna, Meiners, Chad, Milechin, Lauren, Mullen, Julie, Pisharody, Sandeep, Prout, Andrew, Reuther, Albert, Rosa, Antonio, Samsi, Siddharth, Sreekanth, Jon, Stetson, Doug, Yee, Charles, and Michaleas, Peter

Subjects
Networking and Internet Architecture (cs.NI), Social and Information Networks (cs.SI), FOS: Computer and information sciences, Computer Science - Networking and Internet Architecture, Computer Science - Operating Systems, Computer Science - Distributed, Parallel, and Cluster Computing, Operating Systems (cs.OS), Computer Science - Social and Information Networks, Distributed, Parallel, and Cluster Computing (cs.DC)
Abstract

Long range detection is a cornerstone of defense in many operating domains (land, sea, undersea, air, space, ..,). In the cyber domain, long range detection requires the analysis of significant network traffic from a variety of observatories and outposts. Construction of anonymized hypersparse traffic matrices on edge network devices can be a key enabler by providing significant data compression in a rapidly analyzable format that protects privacy. GraphBLAS is ideally suited for both constructing and analyzing anonymized hypersparse traffic matrices. The performance of GraphBLAS on an Accolade Technologies edge network device is demonstrated on a near worse case traffic scenario using a continuous stream of CAIDA Telescope darknet packets. The performance for varying numbers of traffic buffers, threads, and processor cores is explored. Anonymized hypersparse traffic matrices can be constructed at a rate of over 50,000,000 packets per second; exceeding a typical 400 Gigabit network link. This performance demonstrates that anonymized hypersparse traffic matrices are readily computable on edge network devices with minimal compute resources and can be a viable data product for such devices., Accepted to IEEE HPEC, Outstanding Paper Award, 8 pages, 8 figures, 1 table, 70 references. arXiv admin note: text overlap with arXiv:2108.06653, arXiv:2008.00307, arXiv:2203.10230

Published
2022
Full Text
View/download PDF

18. Hypersparse Network Flow Analysis of Packets with GraphBLAS

Author

Trigg, Tyler, primary, Meiners, Chad, additional, Pisharody, Sandeep, additional, Jananthan, Hayden, additional, Jones, Michael, additional, Michaleas, Adam, additional, Davis, Timothy, additional, Welch, Erik, additional, Arcand, William, additional, Bestor, David, additional, Bergeron, William, additional, Byun, Chansup, additional, Gadepally, Vijay, additional, Houle, Micheal, additional, Hubbell, Matthew, additional, Klein, Anna, additional, Michaleas, Peter, additional, Milechin, Lauren, additional, Mullen, Julie, additional, Prout, Andrew, additional, Reuther, Albert, additional, Rosa, Antonio, additional, Samsi, Siddharth, additional, Stetson, Doug, additional, Yee, Charles, additional, and Kepner, Jeremy, additional

Published
2022
Full Text
View/download PDF

19. pPython for Parallel Python Programming

Author

Byun, Chansup, primary, Arcand, William, additional, Bestor, David, additional, Bergeron, Bill, additional, Gadepally, Vijay, additional, Houle, Michael, additional, Hubbell, Matthew, additional, Jananthan, Hayden, additional, Jones, Michael, additional, Keville, Kurt, additional, Klein, Anna, additional, Michaleas, Peter, additional, Milechin, Lauren, additional, Morales, Guillermo, additional, Mullen, Julie, additional, Prout, Andrew, additional, Reuther, Albert, additional, Rosa, Antonio, additional, Samsi, Siddharth, additional, Yee, Charles, additional, and Kepner, Jeremy, additional

Published
2022
Full Text
View/download PDF

20. Python Implementation of the Dynamic Distributed Dimensional Data Model

Author

Jananthan, Hayden, primary, Milechin, Lauren, additional, Jones, Michael, additional, Arcand, William, additional, Bergeron, William, additional, Bestor, David, additional, Byun, Chansup, additional, Houle, Michael, additional, Hubbell, Matthew, additional, Gadepally, Vijay, additional, Klein, Anna, additional, Michnlons, Peter, additional, Morales, Guillermo, additional, Mullen, Julie, additional, Prout, Andrew, additional, Reuther, Albert, additional, Rosa, Antonio, additional, Samsi, Siddharth, additional, Yee, Charles, additional, and Kepner, Jeremy, additional

Published
2022
Full Text
View/download PDF

22. Temporal Correlation of Internet Observatories and Outposts

Author

Kepner, Jeremy, Jones, Michael, Andersen, Daniel, Buluç, Aydın, Byun, Chansup, Claffy, K, Davis, Timothy, Arcand, William, Bernays, Jonathan, Bestor, David, Bergeron, William, Gadepally, Vijay, Grant, Daniel, Houle, Micheal, Hubbell, Matthew, Jananthan, Hayden, Klein, Anna, Meiners, Chad, Milechin, Lauren, Morris, Andrew, Mullen, Julie, Pisharody, Sandeep, Prout, Andrew, Reuther, Albert, Rosa, Antonio, Samsi, Siddharth, Stetson, Doug, Yee, Charles, and Michaleas, Peter

Subjects
Computer Science - Networking and Internet Architecture, Networking and Internet Architecture (cs.NI), Social and Information Networks (cs.SI), FOS: Computer and information sciences, Computer Science - Distributed, Parallel, and Cluster Computing, Computer Science - Social and Information Networks, Distributed, Parallel, and Cluster Computing (cs.DC)
Abstract

The Internet has become a critical component of modern civilization requiring scientific exploration akin to endeavors to understand the land, sea, air, and space environments. Understanding the baseline statistical distributions of traffic are essential to the scientific understanding of the Internet. Correlating data from different Internet observatories and outposts can be a useful tool for gaining insights into these distributions. This work compares observed sources from the largest Internet telescope (the CAIDA darknet telescope) with those from a commercial outpost (the GreyNoise honeyfarm). Neither of these locations actively emit Internet traffic and provide distinct observations of unsolicited Internet traffic (primarily botnets and scanners). Newly developed GraphBLAS hyperspace matrices and D4M associative array technologies enable the efficient analysis of these data on significant scales. The CAIDA sources are well approximated by a Zipf-Mandelbrot distribution. Over a 6-month period 70\% of the brightest (highest frequency) sources in the CAIDA telescope are consistently detected by coeval observations in the GreyNoise honeyfarm. This overlap drops as the sources dim (reduce frequency) and as the time difference between the observations grows. The probability of seeing a CAIDA source is proportional to the logarithm of the brightness. The temporal correlations are well described by a modified Cauchy distribution. These observations are consistent with a correlated high frequency beam of sources that drifts on a time scale of a month., Comment: 8 pages, 8 figures, 2 tables, 59 references; accepted to GrAPL 2022. arXiv admin note: substantial text overlap with arXiv:2108.06653

Published
2022
Full Text
View/download PDF

23. The MIT Supercloud Workload Classification Challenge

Author

Benny J. Tang, Qiqi Chen, Matthew L. Weiss, Nathan C. Frey, Joseph McDonald, David Bestor, Charles Yee, William Arcand, William Bergeron, Chansup Byun, Daniel Edelman, Michael Houle, Matthew Hubbell, Michael Jones, Jeremy Kepner, Anna Klein, Adam Michaleas, Peter Michaleas, Lauren Milechin, Julia Mullen, Andrew Prout, Albert Reuther, Antonio Rosa, Andrew Bowne, Lindsey McEvoy, Baolin LI, Devesh Tiwari, Jiay Gadepally, and Siddharth Samsi

Subjects
FOS: Computer and information sciences, Computer Science - Machine Learning, Computer Science - Distributed, Parallel, and Cluster Computing, Distributed, Parallel, and Cluster Computing (cs.DC), Machine Learning (cs.LG)
Abstract

High-Performance Computing (HPC) centers and cloud providers support an increasingly diverse set of applications on heterogenous hardware. As Artificial Intelligence (AI) and Machine Learning (ML) workloads have become an increasingly larger share of the compute workloads, new approaches to optimized resource usage, allocation, and deployment of new AI frameworks are needed. By identifying compute workloads and their utilization characteristics, HPC systems may be able to better match available resources with the application demand. By leveraging datacenter instrumentation, it may be possible to develop AI-based approaches that can identify workloads and provide feedback to researchers and datacenter operators for improving operational efficiency. To enable this research, we released the MIT Supercloud Dataset, which provides detailed monitoring logs from the MIT Supercloud cluster. This dataset includes CPU and GPU usage by jobs, memory usage, and file system logs. In this paper, we present a workload classification challenge based on this dataset. We introduce a labelled dataset that can be used to develop new approaches to workload classification and present initial results based on existing approaches. The goal of this challenge is to foster algorithmic innovations in the analysis of compute workloads that can achieve higher accuracy than existing methods. Data and code will be made publicly available via the Datacenter Challenge website : https://dcc.mit.edu., Comment: Accepted at IPDPS ADOPT'22

Published
2022
Full Text
View/download PDF

24. The MIT Supercloud Workload Classification Challenge

Author

Tang, Benny J., primary, Chen, Qiqi, additional, Weiss, Matthew L., additional, Frey, Nathan C., additional, McDonald, Joseph, additional, Bestor, David, additional, Yee, Charles, additional, Arcand, William, additional, Bergeron, William, additional, Byun, Chansup, additional, Edelman, Daniel, additional, Houle, Michael, additional, Hubbell, Matthew, additional, Jones, Michael, additional, Kepner, Jeremy, additional, Klein, Anna, additional, Michaleas, Adam, additional, Michaleas, Peter, additional, Milechin, Lauren, additional, Mullen, Julia, additional, Prout, Andrew, additional, Reuther, Albert, additional, Rosa, Antonio, additional, Bowne, Andrew, additional, McEvoy, Lindsey, additional, LI, Baolin, additional, Tiwari, Devesh, additional, Gadepally, Jiay, additional, and Samsi, Siddharth, additional

Published
2022
Full Text
View/download PDF

25. AI-Enabling Workloads on Large-Scale GPU-Accelerated System: Characterization, Opportunities, and Implications

Author

Li, Baolin, primary, Arora, Rohin, additional, Samsi, Siddharth, additional, Patel, Tirthak, additional, Arcand, William, additional, Bestor, David, additional, Byun, Chansup, additional, Roy, Rohan Basu, additional, Bergeron, Bill, additional, Holodnak, John, additional, Houle, Michael, additional, Hubbell, Matthew, additional, Jones, Michael, additional, Kepner, Jeremy, additional, Klein, Anna, additional, Michaleas, Peter, additional, McDonald, Joseph, additional, Milechin, Lauren, additional, Mullen, Julie, additional, Prout, Andrew, additional, Price, Benjamin, additional, Reuther, Albert, additional, Rosa, Antonio, additional, Weiss, Matthew, additional, Yee, Charles, additional, Edelman, Daniel, additional, Vanterpool, Allan, additional, Cheng, Anson, additional, Gadepally, Vijay, additional, and Tiwari, Devesh, additional

Published
2022
Full Text
View/download PDF

28. Supercomputing Enabled Deployable Analytics for Disaster Response

Author

Samuel, Kaira, Kepner, Jeremy, Jones, Michael, Milechin, Lauren, Gadepally, Vijay, Arcand, William, Bestor, David, Bergeron, William, Byun, Chansup, Hubbell, Matthew, Houle, Michael, Klein, Anna, Lopez, Victor, Mullen, Julie, Prout, Andrew, Reuther, Albert, Rosa, Antonio, Samsi, Sid, Yee, Charles, and Michaleas, Peter

Subjects
FOS: Computer and information sciences, Computer Science - Graphics, Computer Science - Databases, Computer Science - Distributed, Parallel, and Cluster Computing, Computer Science - Human-Computer Interaction, Databases (cs.DB), Distributed, Parallel, and Cluster Computing (cs.DC), Computer Science - Multimedia, Graphics (cs.GR), Human-Computer Interaction (cs.HC), Multimedia (cs.MM)
Abstract

First responders and other forward deployed essential workers can benefit from advanced analytics. Limited network access and software security requirements prevent the usage of standard cloud based microservice analytic platforms that are typically used in industry. One solution is to precompute a wide range of analytics as files that can be used with standard preinstalled software that does not require network access or additional software and can run on a wide range of legacy hardware. In response to the COVID-19 pandemic, this approach was tested for providing geo-spatial census data to allow quick analysis of demographic data for better responding to emergencies. These data were processed using the MIT SuperCloud to create several thousand Google Earth and Microsoft Excel files representative of many advanced analytics. The fast mapping of census data using Google Earth and Microsoft Excel has the potential to give emergency responders a powerful tool to improve emergency preparedness. Our approach displays relevant census data (total population, population under 15, population over 65, median age) per census block, sorted by county, through a Microsoft Excel spreadsheet (xlsx file) and Google Earth map (kml file). The spreadsheet interface includes features that allow users to convert between different longitude and latitude coordinate units. For the Google Earth files, a variety of absolute and relative colors maps of population density have been explored to provide an intuitive and meaningful interface. Using several hundred cores on the MIT SuperCloud, new analytics can be generated in a few minutes., Comment: 5 pages, 11 figures, 17 references, accepted to IEEE HPEC 2021

Published
2021
Full Text
View/download PDF

29. Spatial Temporal Analysis of 40,000,000,000,000 Internet Darkspace Packets

Author

Kepner, Jeremy, Jones, Michael, Andersen, Daniel, Buluc, Aydin, Byun, Chansup, Claffy, K, Davis, Timothy, Arcand, William, Bernays, Jonathan, Bestor, David, Bergeron, William, Gadepally, Vijay, Houle, Micheal, Hubbell, Matthew, Klein, Anna, Meiners, Chad, Milechin, Lauren, Mullen, Julie, Pisharody, Sandeep, Prout, Andrew, Reuther, Albert, Rosa, Antonio, Samsi, Siddharth, Stetson, Doug, Tse, Adam, Yee, Charles, and Michaleas, Peter

Subjects
Networking and Internet Architecture (cs.NI), Performance (cs.PF), Social and Information Networks (cs.SI), FOS: Computer and information sciences, Computer Science - Networking and Internet Architecture, Computer Science - Performance, Computer Science - Distributed, Parallel, and Cluster Computing, Computer Science - Social and Information Networks, Distributed, Parallel, and Cluster Computing (cs.DC)
Abstract

The Internet has never been more important to our society, and understanding the behavior of the Internet is essential. The Center for Applied Internet Data Analysis (CAIDA) Telescope observes a continuous stream of packets from an unsolicited darkspace representing 1/256 of the Internet. During 2019 and 2020 over 40,000,000,000,000 unique packets were collected representing the largest ever assembled public corpus of Internet traffic. Using the combined resources of the Supercomputing Centers at UC San Diego, Lawrence Berkeley National Laboratory, and MIT, the spatial temporal structure of anonymized source-destination pairs from the CAIDA Telescope data has been analyzed with GraphBLAS hierarchical hypersparse matrices. These analyses provide unique insight on this unsolicited Internet darkspace traffic with the discovery of many previously unseen scaling relations. The data show a significant sustained increase in unsolicited traffic corresponding to the start of the COVID19 pandemic, but relatively little change in the underlying scaling relations associated with unique sources, source fan-outs, unique links, destination fan-ins, and unique destinations. This work provides a demonstration of the practical feasibility and benefit of the safe collection and analysis of significant quantities of anonymized Internet traffic., 8 pages, 9 figures, 2 tables, 43 references, accepted to IEEE HPEC 2021. arXiv admin note: substantial text overlap with arXiv:2008.00307

Published
2021
Full Text
View/download PDF

31. Eight Bit Quantum Fourier Transform Using the FDTD Method

Author

Jennifer Houle and Dennis M. Sullivan

Subjects
010302 applied physics, Physics, Finite-difference time-domain method, 02 engineering and technology, 021001 nanoscience & nanotechnology, 01 natural sciences, Power (physics), Schrödinger equation, symbols.namesake, Quantum gate, Fourier transform, Dimension (vector space), 0103 physical sciences, symbols, Quantum Fourier transform, 0210 nano-technology, Algorithm, Harmonic oscillator
Abstract

A way of using the Finite Difference Time Domain method is described to simulate the Quantum Fourier Transform, which is an essential component of Shor’s factoring algorithm. This simulation is based on the direct implementation of the time-dependent Schrodinger equation in one dimension. Each bit is simulated as an electron in a harmonic oscillator. The behavior of each quantum gate is simulated by applying a magnetic field in specific orientations for set amounts of time based on the amount of time the electron requires to precess. By using a combination of these quantum gates, it is possible to simulate the behavior of the full Quantum Fourier Transform. An eight bit Quantum Fourier Transform was simulated for this work, but it could easily be expanded to reach higher numbers of bits. Results were compared with computational results and shown to match. Simulations were done in Python without requiring significant computational power.

Published
2021
Full Text
View/download PDF

32. 3D Real-Time Supercomputer Monitoring

Author

Bergeron, Bill, primary, Hubbell, Matthew, additional, Sequeira, Dylan, additional, Williams, Winter, additional, Arcand, William, additional, Bestor, David, additional, Byun, Chansup, additional, Gadepally, Vijay, additional, Houle, Michael, additional, Jones, Michael, additional, Klien, Anna, additional, Michaleas, Peter, additional, Milechin, Lauren, additional, Mullen, Julie, additional, Prout, Andrew, additional, Reuther, Albert, additional, Rosa, Antonio, additional, Samsi, Siddharth, additional, Yee, Charles, additional, and Kepner, Jeremy, additional

Published
2021
Full Text
View/download PDF

33. Vertical, Temporal, and Horizontal Scaling of Hierarchical Hypersparse GraphBLAS Matrices

Author

Kepner, Jeremy, primary, Davis, Tim, additional, Byun, Chansup, additional, Arcand, William, additional, Bestor, David, additional, Bergeron, William, additional, Gadepally, Vijay, additional, Houle, Michael, additional, Hubbell, Matthew, additional, Jones, Michael, additional, Klein, Anna, additional, Milechin, Lauren, additional, Mullen, Julie, additional, Prout, Andrew, additional, Reuther, Albert, additional, Rosa, Antonio, additional, Samsi, Siddharth, additional, Yee, Charles, additional, and Michaleas, Peter, additional

Published
2021
Full Text
View/download PDF

34. Node-Based Job Scheduling for Large Scale Simulations of Short Running Jobs

Author

Byun, Chansup, primary, Arcand, William, additional, Bestor, David, additional, Bergeron, Bill, additional, Gadepally, Vijay, additional, Houle, Michael, additional, Hubbell, Matthew, additional, Jones, Michael, additional, Klein, Anna, additional, Michaleas, Peter, additional, Milechin, Lauren, additional, Mullen, Julie, additional, Prout, Andrew, additional, Reuther, Albert, additional, Rosa, Antonio, additional, Samsi, Siddharth, additional, Yee, Charles, additional, and Kepner, Jeremy, additional

Published
2021
Full Text
View/download PDF

35. Best of Both Worlds: High Performance Interactive and Batch Launching

Author

Peter Michaleas, William Arcand, Julie Mullen, Bill Bergeron, Lauren Milechin, Albert Reuther, David Bestor, Andrew Kirby, Michael Jones, Michael Houle, Jeremy Kepner, Matthew Hubbell, Antonio Rosa, Charles Yee, Siddharth Samsi, Anna Klein, Andrew Prout, Chansup Byun, and Vijay Gadepally

Subjects
FOS: Computer and information sciences, Schedule, business.industry, Computer science, Big data, Preemption, 010103 numerical & computational mathematics, computer.software_genre, Supercomputer, 01 natural sciences, 010305 fluids & plasmas, Scheduling (computing), Idle, Computer Science - Distributed, Parallel, and Cluster Computing, User experience design, 0103 physical sciences, Operating system, Batch processing, Distributed, Parallel, and Cluster Computing (cs.DC), 0101 mathematics, business, computer
Abstract

Rapid launch of thousands of jobs is essential for effective interactive supercomputing, big data analysis, and AI algorithm development. Achieving thousands of launches per second has required hardware to be available to receive these jobs. This paper presents a novel preemptive approach to implement “spot” jobs on MIT SuperCloud systems allowing the resources to be fully utilized for both long running batch jobs while still providing fast launch for interactive jobs. The new approach separates the job preemption and scheduling operations and can achieve 100 times faster performance in the scheduling of a job with preemption when compared to using the standard scheduler-provided automatic preemption-based capability. The results demonstrate that the new approach can schedule interactive jobs preemptively at a performance comparable to when the required computing resources are idle and available. The spot job capability can be deployed without disrupting the interactive user experience while increasing the overall system utilization.

Published
2020
Full Text
View/download PDF

37. Multi-Temporal Analysis and Scaling Relations of 100,000,000,000 Network Packets

Author

Kepner, Jeremy, primary, Meiners, Chad, additional, Byun, Chansup, additional, McGuire, Sarah, additional, Davis, Timothy, additional, Arcand, William, additional, Bernays, Jonathan, additional, Bestor, David, additional, Bergeron, William, additional, Gadepally, Vijay, additional, Harnasch, Raul, additional, Hubbell, Matthew, additional, Houle, Micheal, additional, Jones, Micheal, additional, Kirby, Andrew, additional, Klein, Anna, additional, Milechin, Lauren, additional, Mullen, Julie, additional, Prout, Andrew, additional, Reuther, Albert, additional, Rosa, Antonio, additional, Samsi, Siddharth, additional, Stetson, Doug, additional, Tse, Adam, additional, Yee, Charles, additional, and Michaleas, Peter, additional

Published
2020
Full Text
View/download PDF

38. Accuracy and Performance Comparison of Video Action Recognition Approaches

Author

Hutchinson, Matthew, primary, Samsi, Siddharth, additional, Arcand, William, additional, Bestor, David, additional, Bergeron, Bill, additional, Byun, Chansup, additional, Houle, Micheal, additional, Hubbell, Matthew, additional, Jones, Micheal, additional, Kepner, Jeremy, additional, Kirby, Andrew, additional, Michaleas, Peter, additional, Milechin, Lauren, additional, Mullen, Julie, additional, Prout, Andrew, additional, Rosa, Antonio, additional, Reuther, Albert, additional, Yee, Charles, additional, and Gadepally, Vijay, additional

Published
2020
Full Text
View/download PDF

39. Fast Mapping onto Census Blocks

Author

Kepner, Jeremy, primary, Kipf, Andreas, additional, Engwirda, Darren, additional, Vembar, Navin, additional, Jones, Michael, additional, Milechin, Lauren, additional, Gadepally, Vijay, additional, Hill, Chris, additional, Kraska, Tim, additional, Arcand, William, additional, Bestor, David, additional, Bergeron, William, additional, Byun, Chansup, additional, Hubbell, Matthew, additional, Houle, Michael, additional, Kirby, Andrew, additional, Klein, Anna, additional, Mullen, Julie, additional, Prout, Andrew, additional, Reuther, Albert, additional, Rosa, Antonio, additional, Samsi, Sid, additional, Yee, Charles, additional, and Michaleas, Peter, additional

Published
2020
Full Text
View/download PDF

40. Best of Both Worlds: High Performance Interactive and Batch Launching

Author

Byun, Chansup, primary, Kepner, Jeremy, additional, Arcand, William, additional, Bestor, David, additional, Bergeron, Bill, additional, Gadepally, Vijay, additional, Houle, Michael, additional, Hubbell, Matthew, additional, Jones, Michael, additional, Kirby, Andrew, additional, Klein, Anna, additional, Michaleas, Peter, additional, Milechin, Lauren, additional, Mullen, Julie, additional, Prout, Andrew, additional, Rosa, Antonio, additional, Samsi, Siddharth, additional, Yee, Charles, additional, and Reuther, Albert, additional

Published
2020
Full Text
View/download PDF

41. Benchmarking network fabrics for data distributed training of deep neural networks

Author

Samsi, Siddharth, primary, Prout, Andrew, additional, Jones, Michael, additional, Kirby, Andrew, additional, Arcand, Bill, additional, Bergeron, Bill, additional, Bestor, David, additional, Byun, Chansup, additional, Gadepally, Vijay, additional, Houle, Michael, additional, Hubbell, Matthew, additional, Klein, Anna, additional, Michaleas, Peter, additional, Milechin, Lauren, additional, Mullen, Julie, additional, Rosa, Antonio, additional, Yee, Charles, additional, Reuther, Albert, additional, and Kepner, Jeremy, additional

Published
2020
Full Text
View/download PDF

42. 75,000,000,000 Streaming Inserts/Second Using Hierarchical Hypersparse GraphBLAS Matrices

Author

Kepner, Jeremy, primary, Davis, Tim, additional, Byun, Chansup, additional, Arcand, William, additional, Bestor, David, additional, Bergeron, William, additional, Gadepally, Vijay, additional, Hubbell, Matthew, additional, Houle, Michael, additional, Jones, Michael, additional, Klein, Anna, additional, Michaleas, Peter, additional, Milechin, Lauren, additional, Mullen, Julie, additional, Prout, Andrew, additional, Rosa, Antonio, additional, Samsi, Siddharth, additional, Yee, Charles, additional, and Reuther, Albert, additional

Published
2020
Full Text
View/download PDF

43. Vertical Stacked LEGO-PoL CPU Voltage Regulator.

Author

Baek, Jaeil, Elasser, Youssef, Radhakrishnan, Kaladhar, Gan, Houle, Douglas, Jonathan P., Krishnamurthy, Harish K., Li, Xin, Jiang, Shuai, Sullivan, Charles R., and Chen, Minjie

Subjects
SWITCHED capacitor circuits, VOLTAGE regulators, HIGH voltages, MICROPROCESSOR design & construction, POWER density, MICROPROCESSORS, INTEL microprocessors
Abstract

This article presents a 48–1 V merged-two-stage hybrid-switched-capacitor converter with a linear extendable group operated point-of-load (LEGO-PoL) architecture for ultrahigh-current microprocessors, featuring 3-D stacked packaging and coupled inductors for miniaturized size, fast speed, and vertical power delivery. The architecture is highly modular and scalable. The switched-capacitor circuits are connected in series on the input side to split the high input voltage into multiple stacked voltage domains. The multiphase buck circuits are connected in parallel to distribute the high output current into multiple parallel current paths. It leverages the advantages of switched-capacitor circuits and multiphase buck circuits to achieve soft charging, current sharing, and voltage balancing. The inductors of the multiphase buck converters are used as current sources to soft-charge and soft-switch the switched-capacitor circuits, and the switched-capacitor circuits are utilized to ensure current sharing among the multiphase buck circuits. A 780-A vertical stacked CPU voltage regulator with a peak efficiency of 91.1% and a full load efficiency of 79.2% at an output voltage of 1 V with liquid cooling is built and tested. The switched capacitor circuits operate at 286 kHz and the buck circuits operate at 1 MHz. It regulates output voltage between 0.8 and 1.5 V through the entire 780-A current range. This is the first demonstration of a 48–1 V CPU voltage regulator to achieve over 1-A/mm $^\text{2}$ current density and the first to achieve 1000-W/in $^3$ power density. [ABSTRACT FROM AUTHOR]

Published
2022
Full Text
View/download PDF

44. Optimizing Xeon Phi for Interactive Data Analysis

Author

Michael Jones, Michael Houle, Jeremy Kepner, William Bergeron, Antonio Rosa, Peter Michaleas, William Arcand, Siddharth Samsi, Vijay Gadepally, Julie Mullen, Matthew Hubbell, Charles Yee, Chansup Byun, Albert Reuther, David Bestor, Anne Klein, Lauren Milechin, and Andrew Prout

Subjects
FOS: Computer and information sciences, Computer Science - Performance, Computer science, CPU cache, Ranging, 02 engineering and technology, Parallel computing, Software_PROGRAMMINGTECHNIQUES, Python (programming language), Matrix multiplication, 020202 computer hardware & architecture, Performance (cs.PF), Manycore processor, Computer Science - Distributed, Parallel, and Cluster Computing, 0202 electrical engineering, electronic engineering, information engineering, GNU Octave, Computer Science - Mathematical Software, 020201 artificial intelligence & image processing, Distributed, Parallel, and Cluster Computing (cs.DC), MATLAB, Mathematical Software (cs.MS), computer, Xeon Phi, computer.programming_language
Abstract

The Intel Xeon Phi manycore processor is designed to provide high performance matrix computations of the type often performed in data analysis. Common data analysis environments include Matlab, GNU Octave, Julia, Python, and R. Achieving optimal performance of matrix operations within data analysis environments requires tuning the Xeon Phi OpenMP settings, process pinning, and memory modes. This paper describes matrix multiplication performance results for Matlab and GNU Octave over a variety of combinations of process counts and OpenMP threads and Xeon Phi memory modes. These results indicate that using KMP_AFFINITY=granlarity=fine, taskset pinning, and all2all cache memory mode allows both Matlab and GNU Octave to achieve 66% of the practical peak performance for process counts ranging from 1 to 64 and OpenMP threads ranging from 1 to 64. These settings have resulted in generally improved performance across a range of applications and has enabled our Xeon Phi system to deliver significant results in a number of real-world applications., Comment: 6 pages, 5 figures, accepted in IEEE High Performance Extreme Computing (HPEC) conference 2019

Published
2019
Full Text
View/download PDF

45. Large Scale Parallelization Using File-Based Communications

Author

Michael Jones, Michael Houle, Jeremy Kepner, Siddharth Samsi, Vijay Gadepally, Andrew Prout, Julie Mullen, Peter Michaleas, Antonio Rosa, David Bestor, Bill Bergeron, Chansup Byun, Anna Klein, Albert Reuther, Charles Yee, Matthew Hubbell, and William Arcand

Subjects
FOS: Computer and information sciences, Computer science, Node (networking), Scale (chemistry), Parallel computing, Cluster (spacecraft), Communication architecture, Computer Science - Distributed, Parallel, and Cluster Computing, Overhead (business), Data_FILES, Distributed, Parallel, and Cluster Computing (cs.DC), File system permissions, Secure copy, Protocol (object-oriented programming)
Abstract

In this paper, we present a novel and new file-based communication architecture using the local filesystem for large scale parallelization. This new approach eliminates the issues with filesystem overload and resource contention when using the central filesystem for large parallel jobs. The new approach incurs additional overhead due to inter-node message file transfers when both the sending and receiving processes are not on the same node. However, even with this additional overhead cost, its benefits are far greater for the overall cluster operation in addition to the performance enhancement in message communications for large scale parallel jobs. For example, when running a 2048-process parallel job, it achieved about 34 times better performance with MPI_Bcast() when using the local filesystem. Furthermore, since the security for transferring message files is handled entirely by using the secure copy protocol (scp) and the file system permissions, no additional security measures or ports are required other than those that are typically required on an HPC system.

Published
2019
Full Text
View/download PDF

46. Three Dimensional Time Domain Simulation of the Quantum Magnetic Susceptibility

Author

Ethan Crowell, Mark G. Kuzyk, Jennifer Houle, Sean Mossman, and Dennis M. Sullivan

Subjects
Physics, symbols.namesake, Magnetic domain, Magnetic moment, Quantum electrodynamics, symbols, Expectation value, Magnetostatics, Magnetic susceptibility, Magnetic dipole, Magnetic field, Schrödinger equation
Abstract

A way of using the Finite Difference Time Domain method is described to simulate the magnetic susceptibility of a quantum toroid. This simulation is based on the direct implementation of the time-dependent Schrodinger equation in three dimensions. First, the ground state eigenenergy and eigenstate are found. Next, the expectation value of the quantum magnetic dipole operator is calculated as a function of the applied magnetic field strength with a static magnetic field, and the results are compared with classical results. Then the magnetic dipole moment is calculated with a time-oscillating magnetic field applied. These expectation values are used to calculate the linear and nonlinear magnetic susceptibility of a torus, both without a grating and with a grating to increase irregularities in the shape, by repeating the calculations at various frequencies. The results are consistent with the expected results. This method can be used to calculate the quantum magnetic susceptibility of any structure in order to search for structures with better nonlinear properties.

Published
2019
Full Text
View/download PDF

47. A layered approach to defending against list-linking email bombs

Author

Cristina Houle and Ruchika Pandey

Subjects
Identification (information), CAPTCHA, Email address, Computer science, Server, Pattern recognition (psychology), Denial-of-service attack, Service provider, Computer security, computer.software_genre, computer, Electronic mail
Abstract

Email bombing is a form of Denial of Service (DoS) attack that consists of sending huge volumes of email to one or more email addresses to overflow the mailbox or overwhelm the server where the mailbox is hosted. While this type of attack is not new, we have seen renewed Distributed Denial of Service (DDoS) attacks by email in the past few years via list-linking attacks, where the victim's email address is subscribed to thousands or even tens of thousands of mailing lists. We describe some of the measures that list owners and email service providers have suggested to mitigate such attacks. We then present a case study of a real-world attack, investigating whether characteristics of attack behavior and email attributes surface a workable hypothesis of early detection paradigms. We test our hypothesis on a dataset of hundreds of millions of emails, representing three months of data from 3,000 medium and large organizations. Our technique helped us detect three previously unknown list-linking attacks in the dataset. We determined that anomalous bursts signify meaningful patterns at the user level, but cannot be extrapolated effectively when analyzing bulk enterprise volume data. Effective spam identification in this case needs to consider the unique nature of the messages in the attack, which lends to their exhibiting linguistic similarities, combined with their temporal proximity. We recommend a layered approach to detection and throttling through per-user volume and time-based methodologies paired with phrasal pattern recognition.

Published
2018
Full Text
View/download PDF

48. Design, Generation, and Validation of Extreme Scale Power-Law Graphs

Author

Julie Mullen, Vijay Gadepally, Albert Reuther, Charles Yee, Roger Pearce, Antonio Rosa, William Arcand, David Bestor, Andrew Prout, Matthew Hubbell, Anna Klein, Timothy A. Davis, Michael Jones, Bill Bergeron, Lauren Milechin, Michael Houle, Siddharth Samsi, Hayden Jananthan, Jeremy Kepner, Peter Michaleas, and Geoff Sanders

Subjects
FOS: Computer and information sciences, Theoretical computer science, Discrete Mathematics (cs.DM), Computer science, Computation, 02 engineering and technology, 010501 environmental sciences, 01 natural sciences, symbols.namesake, Kronecker delta, Computer Science - Data Structures and Algorithms, FOS: Mathematics, 0202 electrical engineering, electronic engineering, information engineering, Mathematics - Combinatorics, Data Structures and Algorithms (cs.DS), Graph property, 0105 earth and related environmental sciences, Random graph, Computer Science - Performance, 020207 software engineering, Degree distribution, Graph, Vertex (geometry), Performance (cs.PF), Computer Science - Distributed, Parallel, and Cluster Computing, Bipartite graph, symbols, Distributed, Parallel, and Cluster Computing (cs.DC), Combinatorics (math.CO), Computer Science - Discrete Mathematics, MathematicsofComputing_DISCRETEMATHEMATICS
Abstract

Massive power-law graphs drive many fields: metagenomics, brain mapping, Internet-of-things, cybersecurity, and sparse machine learning. The development of novel algorithms and systems to process these data requires the design, generation, and validation of enormous graphs with exactly known properties. Such graphs accelerate the proper testing of new algorithms and systems and are a prerequisite for success on real applications. Many random graph generators currently exist that require realizing a graph in order to know its exact properties: number of vertices, number of edges, degree distribution, and number of triangles. Designing graphs using these random graph generators is a time-consuming trial-and-error process. This paper presents a novel approach that uses Kronecker products to allow the exact computation of graph properties prior to graph generation. In addition, when a real graph is desired, it can be generated quickly in memory on a parallel computer with no-interprocessor communication. To test this approach, graphs with $10^{12}$ edges are generated on a 40,000+ core supercomputer in 1 second and exactly agree with those predicted by the theory. In addition, to demonstrate the extensibility of this approach, decetta-scale graphs with up to $10^{30}$ edges are simulated in a few minutes on a laptop., Comment: 8 pages, 6 figures, IEEE IPDPS 2018 Graph Algorithm Building Blocks (GABB) workshop

Published
2018
Full Text
View/download PDF

49. Streaming 1.9 Billion Hypersparse Network Updates per Second with D4M

Author

Kepner, Jeremy, primary, Houle, Michael, additional, Jones, Michael, additional, Klein, Anne, additional, Michaleas, Peter, additional, Mullen, Julie, additional, Prout, Andrew, additional, Rosa, Antonio, additional, Yee, Charles, additional, Reuther, Albert, additional, Gadepally, Vijay, additional, Milechin, Lauren, additional, Samsi, Siddharth, additional, Arcand, William, additional, Bestor, David, additional, Bergeron, William, additional, Byun, Chansup, additional, and Hubbell, Matthew, additional

Published
2019
Full Text
View/download PDF

50. Securing HPC using Federated Authentication

Author

Prout, Andrew, primary, Klein, Anna, additional, Michaleas, Peter, additional, Milechin, Lauren, additional, Mullen, Julie, additional, Rosa, Antonio, additional, Samsi, Siddharth, additional, Yee, Charles, additional, Reuther, Albert, additional, Kepner, Jeremy, additional, Arcand, William, additional, Bestor, David, additional, Bergeron, Bill, additional, Byun, Chansup, additional, Gadepally, Vijay, additional, Houle, Michael, additional, Hubbell, Matthew, additional, and Jones, Michael, additional

Published
2019
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results