High Performance Computing: Use of AI and Emerging Technologies in Science
In the HPC universe there is a new normal as far as compute is concerned and the main theme is that during recent times compute is not keeping up. We find ourselves in a situation where we need to continuously adapt to a pace of change dictated by tomorrow’s requirements. In this course, develped in partnership with IEEE Future Directions, we will learn about the leading edge of HPC research that involves understanding how AI can be used in science and the latest developments in heterogeneity. We will see examples of serverless computing that portray the delivery of HPC as a service. We will also describe those technologies that can be used to better deliver AI for science and HPC.
What you will learn:
- Examine how new and old hardware combined in older and newer architectures is generating more compute
- Review the architecture, accelerators, and the programming working together in harmony
- Discuss Heterogeneous Serverless Computing and its architecture
- Consider Uses and Challenges of AI in HPC and science
- Review emerging technologies for AI in HPC and science
This course is part of the following course program:
High Performance Computing: Technologies, Solutions to Exascale Systems, and Beyond
Courses included in this program:
Instructors
Dejan Milojicic
Dr. Milojicic is the author of 80 granted patents, over 240 publications, and two books. Dejan is an IEEE Fellow and the former president of the IEEE Computer Society (2014). He was the editor-in-chief of IEEE Computing Now and IEEE Distributed Systems Online. He envisioned, designed, and taught a Cloud Management class at San Jose State University and has led many industry-government and academic collaborations, such as Open Cirrus for the HPC in the Cloud program and the New Operating System for the Machine program. Today, he works at Hewlett Packard Labs as a distinguished technologist where he leads the research on Future Architectures and is the principal investigator of Heterogeneous Serverless Computing.
Paolo Faraboschi
Dr. Paolo Faraboschi is the author of 60 granted, is the author of 60 granted patents, over 100 publications, and the book “Embedded Computing: a VLIW Approach.” In 2014, Paolo was named IEEE Fellow for contributions to embedded processor architecture and system-on-chip technology. He pioneered low-energy servers with HP’s project Moonshot (2010-2014), helped to engineer the hardware architecture of “The Machine” project (2014-2016), and conducted exascale computing research (2017-2020). Today he is Vice President and HPE Fellow and directs Artificial Intelligence Research at Hewlett Packard Labs.
Publication Year: 2023
ISBN: 978-1-7281-7830-1