|School of Computer Science and Informatics University College Dublin Belfield, Dublin 4 Ireland|
|Research group: Heterogeneous Computing Laboratory (HCL) Director: Dr. Alexey Lastovetsky School of Computer Science and Informatics University College Dublin Belfield, Dublin 4 Ireland||
Founded in 1854, University College Dublin is one of Europe’s leading research-intensive universities. Today UCD is Ireland’s largest and most diverse university with over 30,000 students, drawn from approximately 124 countries. UCD is Ireland’s leader in graduate education with approximately 7,000 graduate students, and almost 2,000 PhD students. Over 50% of UCD undergraduates progress to graduate studies. UCD is home to over 6,000 international students and delivers degrees to over 5,000 students on overseas campuses. The role of UCD within Irish higher education is underscored by the fact that UCD alone accounts for over 30% of international students, over 25% of all graduate students and almost 28% of all doctoral enrolments across the seven Irish universities.
UCD is the national leader in research funding. The university has won over €480m in externally funded research contracts in the last five years. UCD has prioritized five major research themes, one of which is Information, Computation and Communication. UCD is consistently ranked within top 100-150 university in the QS World University ranking.
Research Group Background
The Heterogeneous Computing Laboratory (HCL) is established by its founding Director, Dr. Alexey Lastovetsky, in 2001. The HCL is part of the School of Computer Science and Informatics at the University College Dublin. It has hosted and trained more than 30 researchers to date. The HCL aspires to be one of the world research leaders in the field of high performance heterogeneous computing. Its vision is to propose and develop innovative ideas, models, algorithms and tools aimed at efficient and reliable solution of most challenging scientific and engineering problems on modern highly heterogeneous and hierarchical HPC platforms. The HCL developed and released under GNU Public License (GPL) 12 programming tools and software packages for high performance heterogeneous computing. This software has been downloaded from the HCL website (hcl.ucd.ie) more than 6000 times over last 5 years. More than €2m of research funding has been attracted by the HCL to date.
Dr. Alexey Lastovetsky is the HCL founding Director. His main research interests include algorithms, models and programming tools for high performance heterogeneous computing. He is the author of mpC, the first parallel programming language for heterogeneous networks of computers. He designed HeteroMPI, an extension of MPI for heterogeneous parallel computing (with R. Reddy), and SmartGridSolve, an extension of GridSolve aimed at higher performance of scientific computing on global networks (with T. Brady, et al.). He has contributed into heterogeneous data distribution algorithms (with A. Kalinov, R. Reddy, et al.), proposed and studied realistic performance models of processors in heterogeneous environments, including the functional model and the band model (with R. Reddy and R. Higgins). He also works on communication performance models and optimization of communication operations for heterogeneous and hierarchical networks. He published over 100 technical papers in refereed journals, edited books and proceedings of international conferences. He authored the monographs “Parallel computing on heterogeneous networks” (Wiley, 2003) and “High performance heterogeneous computing” (with J. Dongarra, Wiley, 2009). He serves on the editorial boards of several research journals including Parallel Computing (Elsevier). He helped organize more than 150 international conferences as chair, program chair, member of program committee, keynote speaker, or tutorial presenter. He was elected Vice Chair of the EU COST Action IC0805 “High Performance Computing on Complex Environments” (2009-2013).
Major Interest in Action Scientific Topics
- Performance modelling of heterogeneous and hybrid platforms
- Heterogeneous parallel algorithms
- Data partitioning for heterogeneous and hybrid platforms
- Energy efficient computing
- Algorithms and tools for parallel computing on extreme-scale platforms
- Optimization of real-world data parallel application on heterogeneous, hybrid and extreme-scale systems