synergy.cs.vt.edu www.cs.vt.edu Overview Research Areas At the Synergy Lab, we conduct basic and applied research Systems Systems software and tools; adaptive run-time that explores a breadth of complementary intellectual systems, parallel computing, monitoring and activities that synergistically span all aspects of parallel measurement, emerging platforms in all forms and and distributed computing – from systems software & scales. middleware to tools & applications – in order to accelerate discovery and innovation. Networking Architecture, performance measurement and evaluation, modeling, protocols, routing & Highlights fundamentals and network security. Microsoft Commercial, Featuring Wu Feng Spring 2015, the commercial highlights Feng’s collaboration with Renaissance Accelerating BIG DATA apps, medical researchers at VT and other research centers with such rational drug design, large-scale programs as Computing in the Cloud. genomic search, facilitating CS Best Paper Award at ACM/SPEC ICPE 2013 education, human computer Subramaniam and Feng receive the Best Paper Award for their interaction and for paper on energy-proportional computing at the 4th ACM/SPEC Int’l cancer research. Conference on Performance Engineering, 2013. The Green Computing Book Green Feng publishes The Green Computing Book. This Power modeling and prediction, hardware and book discusses approaches to advance the state of the Computing software controlled power management and art in large-scale green computing. benchmarking on platforms ranging from mobile processors and DSPs Best Poster Award at ACM CODASPY 2015 to supercomputers. Zhang and Feng, in collaboration with Shu and Yao, receive the Best Poster Award for “Rapid Screening of Transformed Data Leaks with Efficient Algorithms and Parallel Computing” at the 5th ACM Conference on Data and Application Security and Privacy in March Research Visit the Pubs of the Synergy Lab 2015. Publications http://synergy.cs.vt.edu/publications.php Paper Selected Among Best in 20 Years of ACM HPDC IEEE/ACM HPDC (1992 – 2012) New York Times: http://nyti.ms/X1DWDs Lin, Ma, Archuleta, Feng, Gardner, Zhang, “MOON: MapReduce On Follow Us in HPCwire: http://bit.ly/ZyX8MM Opportunistic eNvironments,” 19th ACM Int’l Symposium on High- the News Performance Distributed Computing, 2012. Wired: http://bit.ly/ZyXd34 WVTF NPR: http://bit.ly/ZyXuTD Green500 Pioneer to Accelerate Mini-Drone Project Microsoft Commercial: https://youtu.be/7Zw8gKJXgI4 The Synergy Lab, along with collaborators at VT and NCSU, use GPUs to accelerate the CFD simulation of micro-air vehicles. Those who fund our research synergy.cs.vt.edu PPP: Parallel Programming with BIG DATA Analytics Pictures Data is being generated at a rate that outstrips our ability to compute on that data. For The PPP program teaches parallel example, DNA sequencers generate 10-fold programming concepts in a blocks-based more data every 18 months while our compute language that caters to K-12 students, capability only doubles every 18 months. This undergraduate students, or workforce research aims to co-design across multiple professionals to re-train via easily layers (hardware, architecture, algorithms, and understood abstractions. software) to bridge the above gap across SEEC, established July 2014, is The project extends a picture-based disciplines. an interdisciplinary and programming environments, i.e., Snap!, integrative center that spans 20 with blocks providing parallel execution Cloud Computing different departments and institutes across functionality. A further goal is to generate With the simultaneous need for high- Virginia Tech. The vision of the center is to appropriate back-end code for parallel performance computing (HPC) and its democratize parallel and distributed execution on a wide range of platforms, e.g., associated expense, cloud computing offers an computing via the synergistic co-design of from laptops to ORNL's Tiny Titan or even alternative means to access HPC, one without hardware, software, and algorithms to to the Titan supercomputer at ORNL. its exorbitant cost but at the expense of less massively accelerate discovery and control and arguably less performance. This innovation. work seeks to create a new generation of The center "SEECs" to combine the Heterogeneous Parallel Computing, scalable analysis & management software for strengths of research, education, and “client+cloud” environments to deliver HPC for e.g., CPUs+GPUs in HokieSpeed infrastructure at Virginia Tech, along with BIG DATA and scientific computing. social, business, and research opportunities Recent trends have exposed the CPU in society to leverage the power of as a “jack of all (computing) trades, Networking computers for the good of mankind. master of none.” To address this, Our aim is to develop models, architectures and heterogeneous computing systems with protocols for next-generation networks. We Look for future postings to see how using a multiple types of brains have emerged to have focused on performance evaluations of synergistic co-design approach can accelerate herald a new age in supercomputing. existing systems. In doing so we have applications in diverse areas such as cyber Building on our expertise, we address a implemented optimized systems and protocols security, data mining, molecular dynamics, myriad of aspects in accelerator-based as proof of concepts. computational fluid dynamics, robotics, next- parallel computing from systems software to generation DNA sequencing, and middleware and libraries to applications. neuroinformatics. (Much of this work grew out from the NSF Center for High-Performance Reconfigurable Green500 Computing or CHREC.) The Green500 provides rankings of the most The NSF Center for High-Performance energy-efficient supercomputers in the world. Reconfigurable Computing consists of more We raise awareness about power consumption, Contact Information than 30 institutions from academia, industry, promote alternative total cost of ownership Synergy Lab; Torgersen Hall 2050 and government with synergistic interests and performance metrics, and ensure that 620 Drillfield Drive goals in reconfigurable and adaptive computing supercomputers only simulate climate change Virginia Tech for a broad range of missions – from satellites and not create it. Blacksburg, VA 24061 to supercomputers. The Synergy Lab is one of Phone: 540-231-1150 / 540-231-3744 the two labs that lead CHREC activities at the Fax: 540-231-6075 Virginia Tech site. Faculty & Staff Students Those who bring synergy to lab 1. Vignesh 11. Arnab W. Feng M.Gardner 2. Sangeetha 12. Sarunya 3. Xuewen 13. Katy 4. Sajal 14. Varun 5. Islam 15. Anshuman H. Wang H. Trease 6. Ahmed 16. Jing 7. Saravanan 17. Wenjie 8. Kaixi 18. Sharmi 9. Umar 19. Run P. Sathre M. Thomas 10. Konstantinos 20. Xiaodong 21. Da .