In this video from SC14, Arni Mittal and Ethan Oayne from the John Monash Science School in Melbourne embark on a journey to New Orleans to uncover the world of supercomputing.
“Moab Viewpoint is the next generation of Adaptive Computing’s admin portal. This enhanced Web-based graphical user interface enables easy viewing of workload— status, reporting on resource utilization and other system metrics. The Moab Viewpoint Portal plays an instrumental role in ensuring SLAs are met — a key component of Adaptive Computing’s Big Workflow vision — by allowing HPC administrators to maximize uptime and prove services were delivered and resources were allocated fairly.”
“We received an overwhelmingly positive response to the new Moab features during SC14, so we¹re very excited to make the new features generally available. In a competitive computing landscape where enterprises need to accelerate insights, Moab matters,” said Rob Clyde, CEO of Adaptive Computing. “Automating workload workflows is imperative to shorten the timeline to discovery, and this latest version of Moab represents a huge step forward in helping enterprises achieve that. We are excited to reveal our latest innovations and continue driving competitive advantage for our customers.”
“This presentation will highlight the use of GPU ray tracing for visualizing the process of photosynthesis, and GPU accelerated analysis of results of hybrid structure determination methods that combine data from cryo-electron microscopy and X-ray crystallography atom molecular dynamics with all- simulations.”
In this video from SC14, Mark O’Conner from Allinea demonstrates the company’s new Forge software development suite. “A shared, intuitive user interface between the debugger and profiler with a single, shallow learning curve ensures scientific developers and HPC experts alike get the maximum value from your tools investment.”
“Deep neural networks have recently emerged as an important tool for difficult AI problems, and have found success in many fields ranging from computer vision to speech recognition. Training deep neural networks is computationally intensive, and so practical application of these networks requires careful attention to parallelism. GPUs have been instrumental in the success of deep neural networks, because they significantly reduce the cost of network training, which then has allowed many researchers to train better networks. In this talk, I will discuss how we were able to duplicate results from a 1000 node cluster using only 3 nodes, each with 4 GPUs.”