The Libraries is proud to present a spring full of events in our Research Computing Event Series. Learn about graphics processing units, Amazon web services, artificial intelligence, tropical cyclone modeling, and research data management tools in talks and hands-on workshops.
Research Computing events are associated with the new Research Computing Interest Group (RCIG), formed by the NCSU Libraries, College of Sciences IT, and OIT. The RCIG is a forum for information sharing and informal discussion about topics such as high performance computing, supercomputing, cloud computing, and emerging areas such as the Internet of Things (IoT). To receive announcements of upcoming events, join the group.
Events in this series are free and open to NC State faculty and students; advance registration is required. Events in this series are generously supported by Eastman Chemical and the NCSU Friends of the Library.
The spring 2018 Research Computing Event Series schedule is:
“GPU Supercomputing on Campus: More Parallelism Than You Can Handle?”
Wednesday, Feb. 14, 10-11 a.m. in the D. H. Hill Library Multimedia Seminar Center
Can you write a program with 10,000 threads on a single GPU (graphics processing unit)? And orchestrate a program to run on up to 100 computers, each with a GPU? This talk by Frank Mueller, an NC State Computer Science professor, addresses the curious and researchers with advanced simulation needs alike, and gives you an idea how to harness 1 million parallel executions on the Advanced Research Computing (ARC) Supercomputing cluster at NCSU, which can also be used for map/reduce-style cloud computing. Are you up for the challenge?
“Amazon Web Services Research Day”
Monday, Feb. 19, 9:30 a.m.-12:00 p.m. in the D. H. Hill Library ITTC Lab 2
Amazon Web Services (AWS) enables you to increase the speed of research and reduce time-to-results by running research and HPC workloads in the cloud and scaling to larger numbers of parallel tasks Athan would be practical in most on-premises environments. AWS helps by providing CPU, GPU, and FPGA servers on-demand, optimized for specific applications, and allows you to build solutions compliant with HIPAA, FISMA, FedRAMP, PCI, and more.
Spend the morning in a workshop for users who want to learn more about AWS service offerings as well as those looking to optimize their current AWS workflows and incorporate more advanced services to support their research. We will explore AWS core services such as Compute, Storage, Database and Networking. There will be an optional Hands-on-Labs session and 1x1 meeting times available during the afternoon as well.
All levels of AWS experience welcome; contact Jeff Essic for more information: (919) 515-5698, firstname.lastname@example.org
“Artificial Intelligence Using GPUs”
Monday, March 12, 10-11 a.m. in the D. H. Hill Library Multimedia Seminar Center
Data scientists in both industry and academia have been using GPUs for AI and machine learning to make groundbreaking improvements across a variety of applications including image classification, video analytics, speech recognition and natural language processing. In particular, Deep Learning – the use of sophisticated, multi-level “deep” neural networks to create systems that can perform feature detection from massive amounts of unlabeled training data – is an area that has been seeing significant investment and research. With thousands of computational cores and 10-100x application throughput compared to CPUs alone, GPUs have become the processor of choice for processing big data for data scientists.
Kurt Schmidt of NVIDIA joins us to talk about two relatively recent trends that have sparked widespread use of Deep Learning within AI: the availability of massive amounts of training data, and powerful and efficient parallel computing provided by GPU computing. He’ll discuss early adopters of GPU accelerators for machine learning including healthcare, automotive, financial, manufacturing, and many of the largest web and social media companies, along with top tier research institutions in data science and machine learning.
“Why is the numerical modeling of tropical cyclones so challenging?”
Wednesday, March 14, 4-5 p.m. in the D. H. Hill Library Multimedia Seminar Center
Tropical cyclones can influence large areas with high winds, heavy rainfall, and sometimes storm surge. But despite the spatial scale and severity of their impact, they are driven by a relatively small-scale and meteorologically complex power plant.
Dr. Gary Lackmann, a professor in the Marine, Earth, and Atmospheric Sciences department, studies high-impact weather, climate change, and numerical atmospheric modeling. He reviews some of the scientific and computational challenges in accurately representing tropical cyclones in numerical models and describe some of the techniques being used to overcome them. These include accounting for processes spanning a broad range of spatial scales, from convection in the eye wall and spiral bands, to air-sea interactions involving sea spray, waves, and evaporation, to large-scale weather systems that can exert a significant steering influence.
“GlobusWorld Tour at NC State”
Tuesday, March 27, 12:30-5:00 p.m. in the D. H. Hill Library Multimedia Seminar Center
Globus is the de-facto service for research data management. Chief Customer Officer Vas Vasiliadis presents a workshop for system administrators, developers and anyone interested in learning more about Globus. Participants at this hands-on session will learn how the Globus platform simplifies web application development, experiment with new Globus services and APIs, and expand their knowledge of Globus administration features.