Annavaram's research focuses on resilient and energy efficient architectures. He co-invented EPI throttling technique which is now widely used in chip industry for maximizing CMP performance when running parallel workloads. His research publications on prefetching techniques for irregular workloads are amongst the most well cited papers in computer architecture. Most recently his research on using mobile phones for traffic estimation is being transformed into a commercial product offering at Nokia.
Beerel is an Associate Professor within USC Viterbi’s Department of Electrical Engineering, having joined the school’s faculty in 1994 after earning his M.S. and Ph.D. degrees in Electrical Engineering from Stanford University. He is also the Faculty Director of Innotation Studies at the USC Stevens Institute for Innovation. In addition to his academic career, Beerel has consulted for Intel in the area of asynchronous design, Trellisware in the area of communication hardware design, Yuni Networks and AMCC in the area of network chip analysis and verification, and Fulcrum Microsystems in the area of asynchronous design, verification, and CAD.
Breuer's research focuses on the fields of design automation and testing. Among his notable contributions in the area of design automation is the development of both the min-cut and forced-directed placement algorithms. The concepts used here have been incorporated into industrial CAD software, and led to more routable layouts. In the area of testing, his most significant work dealt with the development of a series of tools for test generation, DFT and BIST insertion that paved the way for the emergence of the field of CAD for test. He has had a profound impact on these fields by producing three texts that have been essential reading for engineers, including Digital System Testing and Testable Design, W.H. Freeman and Company, New York, NY, 1990.
Dubois works on parallel architectures while emphasizing multiprocessor memory systems and general-purpose multiprocessor architectures. His seminal research work on cache coherence and logical properties of multiprocessor memory systems in the 1980’s eventually led to the definition of memory consistency models, which today are part of all instruction-set architecture definitions. At the beginning of the 1990’s he developed the first FPGA-based configurable hardware platform to rapidly prototype multiprocessor architectures.
Gupta is an associate professor with USC Viterbi’ Ming Hsieh Department of Electrical Engineering and an associate editor of the IEEE Transactions on Computers. His research interests are in the area of VLSI testing and design and he is currently involved in projects on test and validation of deep submicron circuits, testing multicore systems-on-silicon, and delay testing and diagnosis of digital circuits.
Hwang is internationally recognized as a leading contributor to High-Performance Computer systems and Higher Education in Computer Engineering. His 1993 Advanced Computer architecture textbook is a classic.
Hwang's research has been gradually shifted from Computer architecture to Internet Computing, Web-scale Distributed Systems, and Network Security in the past 15 years. Kai founded the Journal of Parallel and Distributed Computing which has become one of two major flagship Journals in this field.
Jain has interests that span networks and control theory. He has made many contributions to the areas of Network Economics and Game Theory, particularly, Network Market Design. In stochastic control, he has developed an empirical process theory approach to simulation-based methods for solving MDPs. Also, He was one of the first to propose a geographical routing algorithm for wireless ad hoc networks. His research is funded by NSF and he is a recipient of the James H. Zumberge Faculty Research and Innovation Award.
Krishnamachari's works on the design and analysis of algorithms, protocols, and applications for next-generation wireless networks. He is known for his seminal work on several topics concerning both theory and practice in wireless sensor networks: experimental measurements, aggregated data gathering, delay and energy-efficient scheduling, flow optimization, and phase transition phenomena. He is the author of one of the first books on the subject: Networking Wireless Sensors, Cambridge University Press, 2005.
Nazarian joined the computer engineering team at USC Viterbi after earning his Ph.D. in electrical engineering from the school’s Ming Hsieh Department of Electrical Engineering in 2006. His research interests cover the areas of VLSI design and computer-aided-design (CAD) with emphasis on timing/noise analysis and optimization, and noise-aware testing and validation of digital systems. He has about 8 years of industrial experience on computer-aided design, hardware and microcontroller based design, wireless design, and digital system testing. At USC, Shahin teaches classes in computer architecture, computer networks, digital design and VLSI design.
Parker's research focus is on preliminary steps necessary to construct a synthetic brain, following achievements in the area of system-level design tools. Alice is known as one of the founders of the high-level synthesis field that aims to produce working digital systems from behavioral specifications of the algorithms. Alice has made contributions to the representation of digital systems, with a scheme that is now found in commercial products, and to the algorithms and techniques that carry out the design. Almost two decades ago, her research group produced an integrated circuit design of a chip from the functional specification in under 48 hours, using a combination of commercial physical software and system software written by her group.
Pedram's research is in the area of low power electronics and design. He and his students developed many of the techniques for gate-and RT-level power analysis and optimization of digital CMOS circuits, which have now become industry standard. He has also made pioneering technical contributions to system-level power and thermal management, dynamic voltage and frequency scaling in processor chips, display power optimization, and power gating in VLSI circuits. His 1996 co-authored book entitled “Low Power Design Methodologies” continues to be a best-selling book on the subject.
Pinkston's research interests are in the area of communication architectures for parallel computing systems, which include multicore and multiprocessor systems. He is known for his contributions to the design and analysis of interconnection networks and routing algorithms, having published over 100 technical articles and book chapters on related topics. His work has opened new lines of research into the discovery of fully-adaptive routing methods and techniques that increase the efficiency of transporting data packets in interconnection networks. His group developed the publicly-available FlexSim and IRFlexSim network simulators used by many researchers in academia and in industry.
Prasanna's research focuses on the field of parallel and distributed computing, particularly reconfigurable computing. His early contribution, the reconfigurable mesh, which allows connections to be reconfigured during computations, abstracts computations in VLSI, optical, and other technologies, including Nano. This model has been studied intensely by a wide international community since Prasanna’s 1993 paper that introduced it. His recent work has focused on the development of fundamental algorithmic techniques for computing systems based on FPGAs, e.g., tera bit rate routers, deep packet inspection and network security. He has developed number of algorithmic optimizations to develop application accelerators in diverse areas to exploit multi core, many core and GPU based systems. In addition, Prasanna’s recent work is studying challenging problems in energy informatics where in computational techniques are used to optimize energy systems including smart oil field and smart grid. Prasanna's integrated optimization capability that leverages semantic web, machine learning, and social network analysis for the smart oil fields domain is being deployed by energy companies such as Chevron. These techniques are also being applied for real-time demand prediction and load curtailment in smart power grids, by processing and analyzing discrete events and historical information at large scales using public and private Cloud computing platforms. Of further interest are programming abstractions for mapping domain problems to Cloud to achieve reliable and scalable execution based on non-traditional performance metrics.
Psounis works on a variety of computer networks, including the Internet, mobile ad hoc networks, delay and disruptive tolerant networks, sensor networks, mesh networks, peer to peer networks and the web. He was the first to demonstrate formal ways to create miniature network replicas to predict the performance of the larger Internet. Another important contribution has been a systematic study of how to communicate in the presence of network disconnections by exploiting node mobility. More recently, Konstantinos and his collaborators presented the first systematic analysis of the performance of practical random access schemes in the context of multi-hop wireless networks, and rigorously proved that the achievable rate region of such schedulers is quite close to optimal when physical layer and other real-world limitations are taken into consideration.
Puvvada has been teaching design courses in digital logic design, computer architecture, and microprocessors. He takes time to design challenging laboratory exercises and projects which improved design skills among our students. Students learn faster and retain longer when taught through fitting examples. Using FPGA boards, USC CEng students have been able to implement pipelined CPU, cache, FIFOs, and several other simple but effective designs.
Raghavendra’s current research focuses on wireless and sensor networks and delay/disruption tolerant networks. His pioneering work is on energy efficient MAC, routing, and broadcasting protocols. He has worked on energy resource management in a number of sensor network applications including data gathering and situation awareness. He has worked extensively on parallel and distributed systems, interconnection networks, and fault tolerant computing.
Redekopp serves as a teaching faculty primarily in the area of digital design and computer architecture. He has also developed several new and novel courses including a hands-on introductory course to the Computer Engineering and Computer Science major as well as a programming course for electrical and biomedical engineers that focuses on problem solving skills in the context of field-specific problems. Mark has also added hands-on lab components to several courses as well as being involved in the development and implementation of an interdisciplinary capstone design course that integrates students from marketing, fine arts, and engineering schools to develop a new product and its marketing materials. Mark serves on the Division of Engineering Education at USC which is charged with developing novel and innovative curricular changes to the engineering curriculum.
Silvester's primary research area is performance analysis and optimization of computer and communication networks. He did some of the earliest work on multi-hop packet radio networks exploring power control tradeoffs between progress toward the destination and interference with other nodes. He has also worked on traffic analysis and management for both wireless and high speed optical networks. In recent years, he was vice-provost for academic computing and scholarly technology at USC during the decade that the University transitioned into the digital information age as the world-wide-web burst onto the scene. He has played a major role in the development of high performance optical networks for the research and education community at regional, national and international scales through his leadership in CENIC, Internet2, the National LambdaRail, and international network connectivity projects.
From 1969 through 1988 Ung spent half of his time doing test and evaluation of weapons systems at the US Army White Sands Missile Range. Through his work, he was chosen by Society for Comp Simulation to chair three consecutive Aerospace Conferences in 1984, ’86, and ’88. In those conferences he also acted as Editor of the proceedings. He also consulted for other branches of the military and the allies. At USC he has been teaching Real-Time computer systems where students are guided to do hardware-in-the-loop systems based upon his military experience.