Burak Yenier presented this talk at the Stanford HPC Conference. “The UberCloud HPC Experiment brings together over 1000 participants to pave the way for HPC as a Service. Active project teams explore the end-to-end process of accessing remote computing resources in HPC centers and in HPC Clouds to study and overcome the potential roadblocks for industry applications such as CAE, Bio, and Life Sciences.”
In this video from the Adaptive Computing booth at SC13, Brian Andrus from the Naval Postgraduate School presents: Supporting Multiple Researchers Using Moab. To support professors, students, and collaborators, the school’s HPC environment includes a diverse set of x86 processors, accelerators, compilers, and more.
Today Mellanox announced CloudX, a reference architecture for building efficient cloud platforms.
Josh Judd from Warp Mechanics describes how the company delivers Lustre Over ZFS on Linux. “No single technology solves all problems faced in today’s complex world. WARP Mechanics’ philosophy is to customize the many and varied systems into the exact set of solutions required to address the problems. WARP Mechanics leverages tried–and-true technologies from the most advanced systems and removes the complexity, delivering customized turnkey solutions.”
John Gustafson will unveil an all-new approach to parallelism as part of his upcoming keynote at the the Multicore World 2014 conference next week in Auckland, New Zealand. “The Ubox Method, based on a new numerical format that uses metadata to store more information using fewer bits, creates a the richest source of data parallelism since the Monte Carlo method, and redefines what is meant by “high performance.”