Wednesday, March 6 at 7:00pm
Graves Hall, Winants Auditorium and Gallery
263 College Avenue, Holland, MI 49423-3646
On Wednesday, March 6 and Thursday, March 7, 2019, the Natural and Applied Sciences Division will welcome Dr. Valerie Taylor as she headlines the annual Gentile Interdisciplinary Lectureship.
Dr. Taylor is the director of the Mathematics and Computer Science Division at Argonne National Laboratory. Previously, she served at Texas A&M University as head of the computer science and engineering departments, senior associate dean of academic affairs in the College of Engineering, and also a Regents Professor and the Royce E. Wisenbaker Professor in the Department of Computer Science. Prior to her time at Texas A&M, Dr. Taylor was a faculty member in the Electrical Engineering and Computer Science Department at Northwestern University for 11 years. Her research is in the area of high-performance computing, with a focus on performance analysis and modeling of parallel, scientific applications. She is also the chief executive officer and president of the Center for Minorities and People with Disabilities in IT (CMD-IT), and is a Fellow of both the Institute of Electrical and Electronics Engineers, and the Association for Computing Machinery.
Dr. Taylor will lead two sessions during her visit. The first, entitled High Performance Computing: A Case for Performance Analysis, will take place on Wednesdy, March 6 at 7:00pm in Winants Auditorium and is open to the public:
High Performance Computing refers to the aggregation of resources (compute, data, interconnects) to deliver the significant computational power for large-scale problems. Current systems have hundreds of thousands of resources. For example, the Theta supercomputer at Argonne National Laboratory consists of 4,392 nodes, each containing a 64 core Xeon Phi processor, resulting in 281,088 cores. Such machines are used to solve large scale applications in physics or engineering, for which it is important to analyze the performance of the applications to achieve efficient execution. This talk will provide an overview of HPC systems, motivate the need for performance analysis and modeling, and present some research results from the use of the models to improve performance.
A more technical second session geared toward students and faculty entitled Exploring the Tradeoffs Between Performance and Power for Parallel Applications, will be held Thursday, March 7 at 11:00am in the Schaap Auditorium inside the Bultman Student Center:
The demand for computational power continues to drive the deployment of ever-growing parallel systems. Production parallel systems with hundreds of thousands of components are being designed and deployed. Future parallel systems are expected to have millions of processors and hundreds of millions of cores, with power requirements. The complexity of these systems is increasing, with hierarchically configured manycore processors and accelerators, together with a deep and complex memory hierarchy. As a result of the complexity, applications face an enormous challenge in exploiting the necessary parameters for efficient execution. While reducing execution time is still the major objective for high performance computing, future systems and applications will have additional power requirements that represent a multidimensional tuning challenge. To embrace these key challenges, we must understand the complicated tradeoffs among runtime and power, and in some cases resilience strategies. This talk will present our methods and analyses to explore these tradeoffs for parallel applications.
The lectureship is supported by a multi-year grant from the Kavli Foundation of Oxnard, California, to bring prominent scientists to campus.
Please join us for this special time of learning!