GPU Computing
GPU computing provides unique challenges and extreme power, reaching well over a teraflop on relatively inexpensive and ubiquitous desktop supercomputers. Several tools exist to make this power available to python programmers.
Open Space Time
Saturday 8:00 pm - Hanover B
Attending
Ken Seehart
Mike Graham
Yung-Yu Chen
Ian Johnson (enjalot)
Olivier Grisel (ogrisel)
Interests/Topic suggestions
General discussion about GPU computing using python.
Review of Roy Hyunjin Han's talk: Introduction to Parallel Computing on an NVIDIA GPU using PyCUDA
Share experiences of various GPU packages and tools, such as PyCUDA, PyGPU, Copperhead, PyStream, and others.
Consideration of yet another python cuda wrapper, cuda_ct, which Ken Seehart plans to unveil at a poster session: GPU computing and the Next Generation Air Transportation System