posted Wednesday, January 26th, 2011
Computer Science and Electrical Engineering
University of Maryland, Baltimore County
GPU Computing: Past, Present, and Future
Dr. David Luebke
Director of Research, NVIDIA Corporation
1:00-2:15pm Friday, 4 February 2011, ITE 227
Modern GPUs have outgrown their graphics heritage in many ways to emerge as the world's most successful parallel computing architecture. The GPUs that consumers buy to play video games provide a level of massively parallel computation in a single chip that was once the preserve of supercomputers. The raw computational horsepower of these chips has expanded their reach well beyond graphics. Today's GPUs not only render video game frames, they also accelerate astrophysics, video transcoding, image processing, protein folding, seismic exploration, computational finance, radioastronomy, heart surgery, self-driving cars – the list goes on and on.
When thinking about the future of GPUs it is important to reflect on the past. How did this peripheral grow into a processing powerhouse found everywhere from medical clinics to radiotelescopes to supercomputers? Why the graphics card and not the modem, or the mouse? Have GPUs really outgrown graphics and will they thus evolve into pure HPC processors? (hint: no)
This talk is intended as a sort of "state of the union" for GPU computing. I'll briefly cover the dual heritage of GPUs, both in terms of supercomputing and the evolution of fixed function graphics pipelines. I'll discuss "computational graphics", the evolution of graphics itself into a general-purpose computational problem, and how that impacts GPU design and GPU computing. Finally I'll describe the important problems and research topics facing GPU computing practitioners and researchers.
David Luebke helped found NVIDIA Research in 2006 after eight years on the faculty of the University of Virginia. Luebke received his Ph.D. under Fred Brooks at the University of North Carolina in 1998. His principal research interests are GPU computing and real-time computer graphics. Luebke's honors include the NVIDIA Distinguished Inventor award, the NSF CAREER and DOE Early Career PI awards, and the ACM Symposium on Interactive 3D Graphics "Test of Time Award". Dr. Luebke has co-authored a book, a SIGGRAPH Electronic Theater piece, a major museum exhibit visited by over 110,000 people, and dozens of papers, articles, chapters, and patents.
posted Wednesday, January 26th, 2011
Huguens Jean, '03, '11, Ph.D., electrical engineering, and Clifford Muse '11, information systems, returned to Haiti in March 2010, after the devastating January earthquake, to fulfill their grandfather's last request of them. As he was dying of cancer, he asked that at his funeral they celebrate his life and "find the joy." "I had no idea what that meant until we encountered these people in Haiti," said Jean, "These images of life continuing on."
The brothers missed their grandfather's funeral when the earthquake made travel to Port-au-Prince impossible, but they resolved to find a way to honor his memory. The new documentary film Lift Up," co-directed by Jean and UMBC alumnus Philip Knowlton, records their journey back home.
The film will be screened at 8:00pm this on Thursday January 27 in the UMBC Commons Skylight room. Admission is free.
posted Wednesday, January 26th, 2011
Work by Professor Hillol Kargupta and his students and colleagues was recognize as one of the top ten data mining case studies at the 2010 IEEE International Conference on Data Mining. The case study submission cited was described in
Hillol Kargupta, Kakali Sarkar, Michael Gilligan, Parag Namjoshi, Sai Subhash Paruchuru, Thiraphat Pongsudhiraks, and Robert Gilligan, "MineFleet, A Distributed Vehicle Performance Data Stream Mining System"
The MineFleet software was designed for commercial fleet owners and fleet management companies who want the power of advanced predictive vehicle data mining for dramatically reducing the operating costs. It uses powerful data stream mining algorithms for modeling, benchmarking, and monitoring of vehicle health, emissions, driver behavior, fuel-consumption, and fleet characteristics. Work on Minefleet was done at UMBC and Agnik, a a data analytics company for distributed, mobile, and embedded environments based in Columbia MD. You can read more about the Minefleet system here.
posted Tuesday, January 25th, 2011
The North American electric power system has been called the world's largest interconnected machine and is a key part of our national infrastructure. The power grid is evolving to better exploit modern information technology and become more integrated with our cyber infrastructure. This presents unprecedented opportunities for enhanced management and efficiency but also introduces vulnerabilities for intrusions, cascading disruptions, malicious attacks, inappropriate manipulations and other threats. Similar issues are foreseen for other cyber-physical infrastructure systems including industrial control systems, transportation, water, natural gas and waste disposal.
A one-day Smart Grid Cyber Security Conference will be held at UMBC on February 15, hosted by the UMBC Computer Science and Electrical Engineering Department and Maryland Clean Energy Technology Incubator. The conference will be a comprehensive presentation by the National Institute of Standards and Technology regarding an Inter-agency Report 7628 (NISTIR 7628) named Guidelines for Smart Grid Cyber Security which is a critically important document for guiding government, regulatory organizations, industry and academia on Smart Grid cybersecurity. This regional outreach conference is valuable to any organization that is planning, integrating, executing or developing cyber technology for the Smart Grid.
The conference is free, but participants are asked to register in advance to help us organize for the correct number of participants.
A full copy of the 600 page report is available here.
posted Monday, January 24th, 2011
Palanivel Kodeswaran will defend his Ph.D. dissertation, "On the use of context and policies in declarative networked systems", at 10:00am on Monday, 31 January in ITE 325. The research was directed by Professor Anupan Joshi. The defense is open to the public.
Abstract: Managing complex networks while ensuring that certain high level goals such as security are met is a complicated process. This is evidenced by the recent Internet outages caused by operators misconfiguring BGP routers. Clearly, there is a growing need to separate the high level goals/policies from the low level mechanisms that implement the various services. We propose a declarative framework for specifying and enforcing high level policies in networks. The declarative framework offers flexibility in terms of specifying the higher level goals rather than focusing on the lower level mechanisms employed in the implementation, and robustness in terms of recovering from failure. One of the key building blocks of our framework is to allow applications to expose their semantics, thereby allowing the underlying network to exploit the semantics and provide better-than-best-effort service where possible. Our framework employs semantic web languages such as OWL and RDF to formally express application and network specifications, and thereby leverages the inherent reasoning and conflict resolution capabilities of these languages. Once the applications and networks are formally specified in our framework, operators can write adaptation policies to jointly adapt the application and network layers in response to changing network conditions. Our experiments with video over wireless show that the joint adaptation provides higher performance compared to no adaptation as well as application/network layer alone adaptation. Furthermore, the adaptation policies are easy to express in our framework and can be dynamically changed at run time. We also show how our framework can be used to automatically configure BGP routers. High level organizational routing policies can be captured in our framework through appropriate ontological specifications. These specifications which can then be checked for correctness are automatically compiled into appropriate low level BGP configurations by our framework and installed on the routers. Furthermore, the logical basis of our specifications enables reasoning, and routers can engage in an argumentation with their neighbors to diagnose and recover from routing misconfigurations through policy controlled reconfigurations. In cases where the argumentation protocol does not converge or the reconfiguration needed is not permitted by policy, the network administrator is alerted along with a log of the argumentation protocol executed so far, helping in isolating the location and cause of failure.
posted Monday, January 24th, 2011
The CSEE Department's two new Cybersecurity graduate programs begin this semester.
Graduate Certificate in Professional Studies: Cybersecurity Strategy and Policy
This four-course graduate certificate can be completed in a year. Because these courses are not technical (although a technical background may be helpful), this program is available to students with a variety of undergraduate backgrounds. Once accepted into the M.P.S.: Cybersecurity, all four courses count toward that degree. Students may choose to take this certificate by itself, or they may take this certificate and then later complete the master’s.
Master's in Professional Studies: Cybersecurity
This ten-course master’s degree incorporates courses in cybersecurity strategy, policy, and management with more technical, hands-on cybersecurity courses. Rather than having to choose between a management or policy-oriented cybersecurity degree or a completely technical cybersecurity degree, the M.P.S. allows you to customize the mix of courses that best meets your career development needs. The master’s degree may be completed in two to three years, depending on how many courses you take each semester. Classes are offered fall, spring, and summer semesters. M.P.S. courses are offered in a classroom or hybrid (50% online, 50% classroom) format. This approach offers flexibility for students while also providing opportunities for in-person professional networking through a dynamic seminar and research experience.
posted Thursday, January 20th, 2011
UMBC will be the Baltimore host site for the 2011 Global Game Jam, an international computer game making festival taking place this coming weekend, January 28th-30th. This is a 48 hour event, where teams from around the globe work to each develop a complete game over one weekend. The first year had 54 sites in 23 countries. Last year, there were 124 sites in 34 countries. The Baltimore site is open to participants at all skill levels, and it is not necessary to be a UMBC student to register. Thanks to generous support by Next Century there is no registration fee for the UMBC site. Participation will be limited to the first 40 registrants.
The jam will start at 5:00pm on Friday 28 January in the UMBC GAIM Lab, ECS room 005a. At that time, the theme for this year's games will be announced, and we'll brainstorm game ideas and form into teams. There is no need to come as a team: each individual has an equal chance to pitch their game ideas, and you can join the team whose game you like best. Teams will have until 3:00pm on Sunday 30 January to develop their games. We'll have demos of each game and selection of local awards, wrapping up by 5:00pm Sunday.