This time, HiPEAC tries a new publication model. Instead of first publishing your paper in a conference, they do a journal review process for all submitted papers, and those accepted papers will have a chance to get presented in a conference, in addition to the journal publication. Our paper on nested loop compilation for reconfigurable architectures, was accepted for this journal issue. Congrats to those involved in this work!
Special high performance computing seminar by NVIDIA Korea
Friday (2011.11.25) 11:00 (EB1 E204).
1. 11:00~11:45 : GPU based render farm and Cloud computing trend
Speaker: JuSeok Lee
Bio:
- Sir Sales manager at NVIDIA Korea professional solution group
- Director of CSE(Korea Computational science and technology Society)
- 2004.Oct ~2007.Feb : sales at Intel Korea CISD, handheld business unit
- Graduated from Yonsei University, Physics department
2. 11:45~12:30 : CUDA, Parallel Programming with GPU without Tears
Speaker: Hyungon Ryu
Bio:
- Solution Architect at NVIDIA Korea Professional Solution Group
- 2009. Jan. ~ work for NVIDIA Korea
- 2006. Mar. ~ Dept. of Mathematics. Yonsei Univ. (PhD. Candidate)
I have been thinking of the implications of the possible “End of Moore’s Law”. Read a related article, Is Moore’s Party Over?, in Comm of ACM.
Whether the end of Moore’s law is near is a topic of active debate, so I don’t want to delve into that. But most people will agree that things are slowing down when it comes to computing.
If Moore’s law is over, will even the number of cores be flattened? What would that mean to computer scientists, engineers, programmers, students, and teachers? Where will big innovations be made if it’s not in exponential performance growth?