جست‌وجوی استاد و پژوهشگر

George Bosilcaمشاهده پروفایل

استاد پژوهشی

• Distributed Algorithms
• Parallel Programming Paradigms
• High Performance Computing
• +۳ مورد دیگر

George Bosilca is a Research Professor at the University of Tennessee, Knoxville, affiliated with the Department of Electrical Engineering and Computer Science and the Innovative Computing Laboratory. He holds a PhD in Computer Science (University of Paris XI, 2004) and an MS in Math and Computer Science (University of Paris XI, 1999). His research focuses on distributed algorithms, parallel programming paradigms, performance modeling/optimization, and resilience in programming models. He contributes to exascale computing initiatives through projects like PaRSEC and Open MPI. Key research areas include task-based runtimes, MPI standardization for exascale systems, and fault-tolerant distributed computing. His work emphasizes scalable and portable constructs for high-performance applications. Bosilca is involved with the Innovative Computing Laboratory (ICL) and collaborates on projects like the EPEXA ecosystem and Argobots threading framework. Recent publications highlight advancements in asynchronous many-task systems, GPU-accelerated collective operations, and resilience strategies for HPC platforms. His contributions span theoretical frameworks and practical implementations, bridging algorithmic innovation with real-world HPC challenges.

Brandon Luciaمشاهده پروفایل

استاد

• Intermittent Computing
• Energy-harvesting Systems
• Concurrency
• +۷ مورد دیگر

Brandon Lucia is a Full Professor in the Department of Electrical and Computer Engineering at Carnegie Mellon University's College of Engineering. He leads the abstract research group focusing on the intersection of computer architecture, systems, and programming languages. His work bridges theoretical foundations with practical implementations in energy-constrained environments. Lucia's research centers on intermittent computing systems and edge computing in extreme environments. His work on energy-harvesting systems has established fundamental principles for batteryless computing, while his orbital edge computing research pioneers computational intelligence for nanosatellite constellations. These research thrusts address critical challenges in reliability, efficiency, and programmability for systems operating under severe power constraints. His publication record shows a clear evolution from foundational work on intermittent computing models to sophisticated applications in space computing and edge intelligence. Recent publications demonstrate increasing integration of dataflow architectures with energy-harvesting constraints, particularly in satellite constellations where computational resources must be managed across distributed, power-constrained platforms operating in extreme environments. NSF CAREER Award (2017) IEEE TCCA Young Computer Architect Award (2019) Sloan Foundation Fellowship (2021) ASPLOS Best Paper Awards (2018, 2020) OOPSLA Distinguished Paper and Artifact Awards (2015) Lucia actively mentors numerous PhD students including Brad Denby, Zhuo Cheng, and Emily Ruppel, many of whom contribute significantly to his research program. His abstract research group maintains strong industry connections while pursuing fundamental advances in computing systems. The group has developed multiple open-source tools including Legerdemain for program analysis and MultiCacheSim for cache coherence simulation. His laboratory work spans from theoretical foundations of intermittent computing to practical implementations in space systems. Current projects include computational nanosatellite constellations, energy-minimal dataflow architectures, and secure edge computing systems that operate reliably despite frequent power failures.

Dominik Huberمشاهده پروفایل

پژوهشگر

• Dynamic Resource Management in High-Performance Computing (HPC)
• Parallel & Distributed Programming Models
• Hardware-aware Programming
• +۶ مورد دیگر

Dominik Huber is a Ph.D. candidate and researcher at the Technical University of Munich , affiliated with the Chair of Computer Architecture & Parallel Systems . His work focuses on Dynamic Resource Management in High-Performance Computing (HPC) , with expertise in Parallel & Distributed Programming Models and Hardware-aware programming . He has actively contributed to teaching courses like Parallel Programming Systems and Advanced Computer Architecture . His research emphasizes adaptive resource allocation in hybrid HPC clusters, leveraging technologies such as MPI Sessions , PMIx , and frameworks like LAIK and XBraid . Recent projects include the DynRes software suite for dynamic resource management and collaborations on quantum-HPC integration. Huber has advised students on topics ranging from Dynamic Resource Management in Charm++ to CI Systems for HPC Software , and his publications address challenges in malleability, scheduling, and power-constrained environments. Current affiliations include participation in the SEANERGYS (EuroHPC) and PlasmaPEPS projects.

• Computer Architecture
• Programming Models
• Parallel Computing
• +۴ مورد دیگر

Daniel Jimenez Gonzalez is a faculty member at the Universitat Politècnica de Catalunya (UPC), affiliated with the Department of Computer Architecture within the School of Informatics (FIB). He is an active researcher in programming models and high-performance computing, contributing extensively to the field through publications, projects, and academic supervision. Research Interests: His work focuses on Computer Architecture , Programming Models , and Parallel Computing . He investigates task-based programming, runtime systems, compiler optimizations, and performance modeling for multi-core and heterogeneous architectures. His research addresses challenges in scalability, energy efficiency, and resilience in modern computing environments. His recent publications show a consistent trend in task-based programming models , runtime scheduling , and performance optimization across diverse architectures, including NUMA and GPU-accelerated systems. The work spans from theoretical modeling to practical implementation, often targeting real-world HPC applications. Scientific Awards: No awards explicitly mentioned in the provided text. Advising and Grants: While specific students are not listed, his involvement in doctoral theses and R&D projects suggests an active role in mentoring graduate students. He has participated in both competitive and non-competitive R&D+i projects, indicating grant acquisition and project leadership experience in areas related to programming models and computer architecture. Labs and Teams: He is a member of the UPC PM - Programming Models research group, which focuses on the design, analysis, and optimization of modern programming paradigms for high-performance systems.

• Parallel Computing
• High-Performance Computing
• Computer Architecture
• +۳ مورد دیگر

Prof. Laxmikant V. Kale is a Research Professor and Paul and Cynthia Saylor Professor Emeritus at the University of Illinois at Urbana-Champaign's Department of Computer Science within the College of Engineering. He specializes in parallel computing, high-performance systems, and scientific applications. His research focuses on runtime systems like Charm++, NAMD, and OpenAtom, emphasizing scalability and efficiency in distributed environments. Education: B. Tech., Electrical Engineering, Banaras Hindu University (1977) M.E., Computer Science, Indian Institute of Science (1979) M.S. & Ph.D., Computer Science, SUNY Stony Brook (1985) Research Interests: Parallel programming models, adaptive runtime systems, distributed and GPU computing, and applications in molecular dynamics, computational biology, and astrophysics. His work bridges theory and practice, aiming to optimize performance for exascale systems. Awards: Gordon Bell Award (2002) IEEE Fellow (2010) Sidney Fernbach Award (2012) HPCC Challenge Award (2011) Grants & Projects: NIH-funded NAMD development, NSF projects on ParaTreet and OPEN ATOM, DOE's XPACC Center, and collaborations with NCSA. He advises students on HPC and molecular dynamics, with notable advisees including Maya Taylor and Jessica Williams. Labs & Teams: Director of the Parallel Programming Laboratory, leading Charm++ and NAMD development. Collaborates with teams on OpenAtom for ab initio simulations and EpiSimdemics for disease modeling.

• High-Performance Computing
• Parallel Programming
• GPU Computing
• +۳ مورد دیگر

assoz. Prof. Dipl.-Ing. Dr. Peter Thoman is an Associate Professor in the Department of Computer Science at the University of Innsbruck. His work focuses on high-performance computing, parallel programming models, and GPU-accelerated systems. He leads research in areas such as SYCL standard implementations, distributed computing optimizations, and energy-efficient algorithms for exascale applications. His research interests include task-based parallelism, heterogeneous computing architectures, and scalable runtime systems. Notable contributions include the Celerity API for accelerator clusters and the LIGATE drug discovery platform. He is affiliated with the University of Innsbruck’s technical research groups and collaborates on projects like the AllScale framework. Dr. Thoman’s teaching includes courses on parallel programming and high-performance computing systems. He is a member of the Department’s leadership team under Prof. Thomas Fahringer.

Dirk Pflügerمشاهده پروفایل

استاد

• Scientific Computing
• High-Performance Computing (HPC)
• Parallel and Distributed Systems
• +۳ مورد دیگر

Dirk Pflüger is a Professor at the University of Stuttgart's Institute of Parallel and Distributed Systems, within the Faculty of Computer Science, Electrical Engineering and Information Technology. His research focuses on high-performance computing (HPC), parallel and distributed systems, and sparse grids. He has led projects in astrophysical simulations, machine learning applications, and uncertainty quantification. Notable contributions include developing scalable algorithms for exascale computing using HPX, Kokkos, and SYCL frameworks. His expertise spans distributed computing architectures, task-based parallel programming, and interdisciplinary applications in astrophysics and medical AI. Recent work includes optimizing hyperparameter tuning, simulating stellar mergers, and enhancing blood glucose prediction models using deep reinforcement learning. Pflüger's research emphasizes performance portability, fault tolerance, and cross-platform collaboration. He has contributed to open-source tools like PLSSVM and hws, which address hardware monitoring and GPU acceleration challenges. His work bridges theoretical advancements with practical implementations for real-world computational problems.

• Computer Science
• Parallel Computing
• Data Stream Processing
• +۳ مورد دیگر

Dr. Jakub Yaghob is a researcher at the Faculty of Mathematics and Physics , Charles University , specializing in computer science and parallel computing. He teaches advanced programming topics including Compiler Principles , Parallel Programming , and Cloud Computing . Research Interests : Parallel data stream processing, virtualization technologies, semantic web infrastructures, and performance optimization Teaching : Advanced C++ programming, virtualization administration, and computer systems architecture Technical Expertise : Design of parallelization frameworks, astrophysical data analysis, and hybrid CPU-GPU systems His publications focus on: Optimizing stream data processing across distributed architectures Developing domain-specific languages like Bobolang Performance evaluation in educational programming contexts Applications of parallel computing in astrophysics

• Robotics
• Optimal Control
• Model Predictive Control
• +۵ مورد دیگر

Wilm Decré is an Associate Professor at KU Leuven, holding a prestigious BOF (Bijzonder OnderzoeksFonds) position in the Department of Mechanical Engineering within the Faculty of Engineering Science. He is a key member of the MECO Research Team, where he leads research in robotics, control systems, and motion planning. His work bridges theoretical control algorithms with practical robotic applications, particularly in optimal control and motion planning. His research interests include: Model Predictive Control (MPC) and optimal control for robotics Time-optimal motion planning and trajectory optimization Non-holonomic path planning algorithms Constraint-based robot programming Human-robot collaboration and safety Real-time implementation of control algorithms Prof. Decré's recent publications demonstrate significant contributions to making advanced control techniques computationally feasible for real-world deployment. His work addresses the gap between theoretical optimality and practical implementation constraints, with applications spanning industrial robotics, autonomous vehicles, and collaborative systems. His research shows a progression from theoretical foundations to increasingly practical applications with real-world impact. His notable scientific contributions include: ASAP-MPC: An asynchronous update scheme for online motion planning Efficient techniques for reducing computational complexity in optimal control Practical approaches for time-optimal MPC implementation Accelerated Reeds-Shepp algorithms for mobile robot path planning Integration of MPC with constraint-based robot programming for unstructured environments Prof. Decré actively supervises PhD students and has developed strong collaborations both within KU Leuven and internationally. His research is supported by the BOF position and appears to involve multiple industry partnerships given the applied nature of many projects. He maintains an active publication record in top robotics and control venues including IEEE Transactions on Robotics, IEEE Robotics and Automation Letters, and major conferences like ICRA and IROS. His laboratory, part of the MECO Research Team, focuses on experimental validation of control algorithms using various robotic platforms including drone systems, mobile robots, and industrial manipulators. The lab emphasizes both theoretical rigor and practical applicability of control solutions, with recent work showing validation on quadcopters navigating cluttered environments and scale model truck-trailers in structured lab environments.

• High-Performance Computing
• Parallel Computing
• Distributed Systems
• +۶ مورد دیگر

Marc Snir is a Professor at the University of Illinois’s Siebel School of Computing and Data Science. He has led significant research contributions in high-performance parallel computing, including work on the Message Passing Interface (MPI) and IBM’s SP scalable parallel system. As Department Head from 2001–2007, he oversaw the transition to the Siebel Center and expanded the department’s capabilities. He later served as the first director of the Illinois Informatics Institute, chief software architect for the Blue Waters supercomputer, and co-director of the Universal Parallel Computing Research Center (UPCRC). His research focuses on parallel computing systems, fault resilience, and I/O optimizations. He has been recognized with the 2014 Distinguished Alumni Service Award. His work spans exascale computing, distributed systems, and machine learning applications in HPC. He has contributed to projects like Argo (an exascale OS/runtime), Aluminum (a GPU-aware communication library), and LCI (Lightweight Communication Interface). His research emphasizes improving scalability, energy efficiency, and reliability in high-performance systems. He has advised numerous students (names not listed here) and led teams in advancing HPC tools and methodologies. His involvement in initiatives like UPCRC and the Blue Waters project underscores his commitment to bridging theoretical research and practical applications in computing.

• High-Performance Computing
• Parallel Computing
• Distributed Systems
• +۴ مورد دیگر

George Bosilca is a Professor at the University of Tennessee, Knoxville, specializing in high-performance computing and parallel systems. With over two decades of research contributions, he has established himself as a leading expert in task-based runtime systems, distributed computing, and MPI implementations. His research focuses on developing and optimizing task-based runtime systems for extreme-scale computing environments, with particular emphasis on fault tolerance, performance optimization, and scalability. Bosilca's work spans multiple domains including scientific computing, climate modeling, and deep learning applications. He has made significant contributions to the PaRSEC runtime system and has extensively researched MPI optimization techniques for modern HPC architectures. The trend in Bosilca's recent publications demonstrates a strong focus on addressing challenges in exascale computing, including fault tolerance in distributed systems, GPU acceleration for scientific workloads, and energy-efficient computing techniques. His work bridges theoretical computer science with practical implementations for real-world scientific applications across various domains. Bosilca maintains extensive collaborations with leading researchers in the HPC community, most notably with Jack J. Dongarra (103 co-publications), Aurelien Bouteiller (60 co-publications), and Thomas Hérault (54 co-publications). These collaborations have resulted in numerous publications at top-tier conferences including SC, IPDPS, and EuroMPI, as well as in prestigious journals such as IEEE Transactions on Parallel and Distributed Systems and the International Journal of High Performance Computing Applications.

• Network-on-Chip (NoC)
• Embedded Systems
• Fault Tolerant Design
• +۴ مورد دیگر

Martin Radetzki is a full Professor at the Institute for Computer Architecture and Parallel Systems (University of Stuttgart) , specializing in Embedded Systems . His work focuses on network-on-chip (NoC) design, fault tolerance, memory optimization, and simulation frameworks. Key research areas: NoC synthesis, deadlock-free routing, performability analysis, and power-efficient memory subsystems Recent publications emphasize integer linear programming frameworks for co-designing floorplanning and routing, chiplet-based systems , and machine learning-enabled performance evaluation His methodologies address cross-layer challenges in NoC design, combining formal optimization with practical implementation for heterogeneous processing elements. Collaborative projects include fault resilience analysis, parallel simulation techniques, and memory allocation strategies for SoCs. Dr. Radetzki supervises research with students like Shuang Liu and Manuel Strobel , contributing to IEEE Transactions on Computers , ACM TECS , and conferences such as DATE and MCSoC . Current work explores optimal routing topologies for emerging chip architectures.

Andrew Lumsdaineمشاهده پروفایل

عضو هیئت علمی

• High Performance Computing
• Scalable Graph Algorithms
• Computational Photography
• +۵ مورد دیگر

Andrew Lumsdaine is the Chief Scientist at the Northwest Institute for Advanced Computing , a dual appointee between the University of Washington and the Pacific Northwest National Laboratory (PNNL). He holds the title of Affiliate Professor in the Paul G. Allen School of Computer Science and Engineering at UW and serves as a Laboratory Fellow in PNNL's Applied Mathematics, Computing, and Data Division. His research spans foundational and applied aspects of High Performance Computing , focusing on scalable graph algorithms, computational photography, and runtime systems for distributed-memory architectures. Education : Not explicitly stated in the provided text. Lumsdaine's work addresses critical challenges in parallel and distributed computing , including synchronization avoidance, communication optimization, and domain-specific language design for graph analytics. He has led projects like GraphPack (NSF-sponsored) and contributed to DARPA's HIVE program through the HAGGLE software development kit. His publications highlight innovations in light field imaging , GPU programming models , and graph algorithm abstractions . Notable collaborations include the GraphBLAS standardization effort and development of tools for checkpoint/restart fault tolerance. His research has been presented at leading conferences like SC , IPDPS , and Eurographics . Lumsdaine actively seeks collaborators and advises students/postdocs through projects listed on his research page.