Department News
An Introduction to the High Performance Computing
Seminar Date
2005-04-18
Author
임아주
Date
2005-04-16
Views
1861
1. 제 목 : An Introduction to the High Performance Computing (HPC)
2. 연 사 : Dr. Junwoo Lim
Sr. Scientific Specialist
Pittsburgh Supercomputing Center
Carnegie Mellon University
3. 일 시 : 2005년 4월 18일(월) 오후 5시
4. 장 소 : 301동 1512호 세미나실
5. 내 용 : Computational demands are growing in virtually every area of scientific research and advanced engineering, from space, climate, and genetics research, to automobile crash simulations and materials modeling, just to name a few. Thanks to the rapid evolution of hardware and software technologies, along with the innovations in numerical analysis tools, computational science has made dazzling advances for the last few decades. So many new technologies and new buzz-words have come and gone, yet the list of such ‘must-know’ items seems to continue to grow even longer. Indeed, the high performance computing (HPC) is now about to blooming. In the past, only classified government agencies or a few selected research institutes can access these HPC resources. Now most colleges have at least one or more HPC clusters and many companies use HPC to improve their business activities.
In this introductory talk, I would like to briefly recollect some footprints of HPC history and hope to share some thoughts on the on-going challenges and outlooks. Some key items below will also be discussed:
-Top 500 list
-Why and how a supercomputer is built?
-Supercomputing, cluster computing, grid computing, and meta-computing
-Scalar architecture, vector architecture, and parallel architecture
-Why go parallel?
-Parallelization with MPI, PVM, or OpenMP
-Optimization and debugging
-Check-pointing and data mining (especially over the ‘grid’)
-Scalability issues (hardware, software, and human-ware)
-New tools for parallel computing
-From Tera-computing to Peta-computing: what are the issues?
-Status of international collaborations and competitions: Building a community
If time permits, a sneak preview of the new 10-teraflops Cray XT-3 supercomputer in Pittsburgh Supercomputing Center will be provided.
6. 문 의 : 기계항공공학부 최 해 천 교수 (☏ 880-8361)
2. 연 사 : Dr. Junwoo Lim
Sr. Scientific Specialist
Pittsburgh Supercomputing Center
Carnegie Mellon University
3. 일 시 : 2005년 4월 18일(월) 오후 5시
4. 장 소 : 301동 1512호 세미나실
5. 내 용 : Computational demands are growing in virtually every area of scientific research and advanced engineering, from space, climate, and genetics research, to automobile crash simulations and materials modeling, just to name a few. Thanks to the rapid evolution of hardware and software technologies, along with the innovations in numerical analysis tools, computational science has made dazzling advances for the last few decades. So many new technologies and new buzz-words have come and gone, yet the list of such ‘must-know’ items seems to continue to grow even longer. Indeed, the high performance computing (HPC) is now about to blooming. In the past, only classified government agencies or a few selected research institutes can access these HPC resources. Now most colleges have at least one or more HPC clusters and many companies use HPC to improve their business activities.
In this introductory talk, I would like to briefly recollect some footprints of HPC history and hope to share some thoughts on the on-going challenges and outlooks. Some key items below will also be discussed:
-Top 500 list
-Why and how a supercomputer is built?
-Supercomputing, cluster computing, grid computing, and meta-computing
-Scalar architecture, vector architecture, and parallel architecture
-Why go parallel?
-Parallelization with MPI, PVM, or OpenMP
-Optimization and debugging
-Check-pointing and data mining (especially over the ‘grid’)
-Scalability issues (hardware, software, and human-ware)
-New tools for parallel computing
-From Tera-computing to Peta-computing: what are the issues?
-Status of international collaborations and competitions: Building a community
If time permits, a sneak preview of the new 10-teraflops Cray XT-3 supercomputer in Pittsburgh Supercomputing Center will be provided.
6. 문 의 : 기계항공공학부 최 해 천 교수 (☏ 880-8361)
