Surendra Byna is a Research Scientist in the Scientific Data Management Group at Lawrence Berkeley National Lab (LBNL). His research interests are in computer architecture and parallel computing. Specifically, he interested in optimizing data access performance for parallel computing and in utilizing heterogeneous computing power. He is also interested in energy efficient parallel computing.
Before joining LBNL, Byna was a researcher at NEC Labs America, where he was a part of the Computer Systems Architecture Department and was involved in the Heterogeneous Cluster Computing project. Prior to that, he was a Research Assistant Professor in the Department of Computer Science at Illinois Institute of Technology (IIT) and a Guest Researcher at the Math and Computer Science division of the Argonne National Laboratory, as well as a Faculty Member of the Scalable Computing Software Laboratory at IIT.
- Ph.D., Computer Science, Illinois Institute of Technology, Chicago, Fall 2006
- M.S., Computer Science, Illinois Institute of Technology, Chicago, Spring 2001
- B.Tech., Electronics and Telecommunication Engineering, Jawaharlal Nehru Tech Univ. (JNTU), Anantapur, India, 1997
Y. Yin, S. Byna, H. Song, X.-H. Sun, and R. Thakur, "Boosting Application-Specific Parallel I/O Optimization Using IOSIG", IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGrid), May 13, 2012,
Spyros Blanas, Wu, Byna, Dong, Arie Shoshani, "Parallel data analysis directly on scientific file formats", SIGMOD Conference, 2014, 385-396,
E. Wes Bethel, Prabhat, Suren Byna, Oliver Rübel, K. John Wu, and Michael Wehner, "Why High Performance Visual Data Analytics is both Relevant and Difficult", Proceedings of Visualization and Data Analysis 2013, IS&T/SPIE Electronic Imaging 2013, San Francisco, CA, USA, SPIE, February 2013, LBNL LBNL-6063E,
- Download File: LBNL-6063E.pdf (pdf: 3.6 MB)
Surendra Byna, Jerry Chou, Oliver Rübel, Prabhat, Homa Karimabadi, William S. Daughton, Vadim Roytershteyn, E. Wes Bethel, Mark Howison, Ke-Jou Hsu, Kuan-Wu Lin, Arie Shoshani, Andrew Uselton, and Kesheng Wu, "Parallel I.O, Analysis, and Visualization of a Trillion Particle Simulation", SuperComputing 2012 (SC12), Salt Lake City, Utah, November 2012,
Prabhat, Oliver Rübel, Surendra Byna, Kesheng Wu, Fuyu Li, Michael Wehner and E. Wes Bethel, "TECA: A Parallel Toolkit for Extreme Climate Analysis", Procedia Computer Science, Proceedings of the International Conference on Computational Science, ICCS 2012, Presented at Third Worskhop on Data Mining in Earth System Science (DMESS 2012), Omaha, Nebraska, June 2012, 9:866–876, LBNL 5352E, doi: 10.1016/j.procs.2012.04.093
We present TECA, a parallel toolkit for detecting extreme events in large climate datasets. Modern climate datasets expose parallelism across a number of dimensions: spatial locations, timesteps and ensemble members. We design TECA to exploit these modes of parallelism and demonstrate a prototype implementation for detecting and tracking three classes of extreme events: tropical cyclones, extra-tropical cyclones and atmospheric rivers. We process a modern TB-sized CAM5 simulation dataset with TECA, and demonstrate good runtime performance for the three case studies.
E. W. Bethel, Surendra Byna, Jerry Chou, Cormier-Michel, Cameron G. R. Geddes, Howison, Fuyu Li, Prabhat, Ji Qiang, Ruebel, Rob D. Ryne, Michael Wehner, Wu, "Big Data Analysis and Visualization: What Do LINACS Tropical Storms Have In Common?", 11th International Computational Accelerator Physics ICAP 2012, Germany, 2012,
Suren Byna, Prabhat, Michael F. Wehner and Kesheng Wu, "Detecting Atmospheric Rivers in Large Climate Datasets", Proceedings of the 2nd International Workshop on Petascale Data Analytics: Challenges, and Opportunities (PDAC-11/ Supercomputing11/ ACM/IEEE), November 14, 2011, Seattle, Washington, 2011, doi: 10.1145/2110205.2110208
Extreme precipitation events on the western coast of North America are often traced to an unusual weather phenomenon known as atmospheric rivers. Although these storms may provide a significant fraction of the total water to the highly managed western US hydrological system, the resulting intense weather poses severe risks to the human and natural infrastructure through severe flooding and wind damage. To aid the understanding of this phenomenon, we have developed an efficient detection algorithm suitable for analyzing large amounts of data. In addition to detecting actual events in the recent observed historical record, this detection algorithm can be applied to global climate model output providing a new model validation methodology. Comparing the statistical behavior of simulated atmospheric river events in models to observations will enhance confidence in projections of future extreme storms. Our detection algorithm is based on a thresholding condition on the total column integrated water vapor established by Ralph et al. (2004) followed by a connected component labeling procedure to group the mesh points into connected regions in space. We develop an efficient parallel implementation of the algorithm and demonstrate good weak and strong scaling. We process a 30-year simulation output on 10,000 cores in under 3 seconds.
Mehmet Balman, Suredra Byna, "Open Problems in network-aware data management in exa-scale computing and terabit networking era", In Proceedings of the First international Workshop on Network-Aware Data Management, in conjunction with ACM/IEEE international Conference For High Performance Computing, Networking, Storage and Analysis, 2011, Seattle, WA, November 11, 2011, LBNL 6176E, doi: http://dx.doi.org/10.1145/2110217.2110229
- Download File: ndm12ppbalman.pdf (pdf: 141 KB)
Accessing and managing large amounts of data is a great challenge in collaborative computing environments where resources and users are geographically distributed. Recent advances in network technology led to next-generation high- performance networks, allowing high-bandwidth connectiv- ity. Efficient use of the network infrastructure is necessary in order to address the increasing data and compute require- ments of large-scale applications. We discuss several open problems, evaluate emerging trends, and articulate our per- spectives in network-aware data management.