Computer Science Seminar

• October 21, 2022
• 12:00 PM - 2:00 CST
• Doyle Center Lower-Level space
• Mohammed Abuhamad, mabuhamad@luc.edu
• Open to the public.
• Add to calendar
• Details

  Hi everyone,

  Next Friday, October 21, the Computer Science Department will be continuing with our CS Seminar Series featuring various exciting speakers covering topics related to the field (see below for next Friday¿s speaker bios and talk abstracts). Please RSVP your attendance to our event by next Wednesday, October 19, to enjoy a meal, social, and two talks by Dr. Nicholas Hopper and Dr. Venkatram Vishwanath. Thank you, and we¿ll see you there! -Marina

  *Note: This talk will be held in the Doyle Center Lower-Level space (a change of venue from our previous talks).

  Schedule:
  12:00 - 12:40 PM - Lunch and Social
  12:40 - 01:20 PM - Talk by Dr. Nicholas Hopper (University of Minnesota)
  01:20 - 02:00 PM - Talk by Dr. Venkatram Vishwanath (Argonne National Laboratory)

  Speakers and Talks:

  --- Defending against Internet traffic analysis
      Talk By: Dr. Nicholas Hopper (University of Minnesota)

  Abstract: Our personal data, business transactions, and even government secrets are all frequently communicated over the Internet. Fortunately, we have seen enormous progress over the past 10 years in the use of encryption to protect these communications. However, encryption alone cannot fully protect the confidentiality of the communications, often leaving the metadata - timing, length, sender and recipient of messages - unprotected. In this talk, I'll discuss recent attacks on systems intended to protect the metadata of Internet communications, and improved defenses against these attacks.

  Speaker Bio: Nicholas Hopper is a Professor and Director of Undergraduate Studies in Computer Science at the University of Minnesota. He received a B.A. from the University of Minnesota in 1999 and a Ph.D. in Computer Science from Carnegie Mellon University in 2004. His research interests include online privacy, applied cryptography, and computer security.

  
  --- Understanding the Efficacy of Emerging AI Accelerators for Science
      Talk by Dr. Venkatram Vishwanath (Argonne National Laboratory)

  Abstract: Scientific applications are increasingly adopting Artificial Intelligence (AI) techniques. We are witnessing emergence of novel hardware accelerators, such as data-flow accelerators and wafer-scale engines, to efficiently run AI-driven science applications. With a wide diversity in the hardware architectures and software stacks of these systems, understanding and comparing the performance of these systems is of paramount importance. The state-of-the-art in the evaluation of deep learning workloads primarily focuses on CPUs and GPUs. In this talk, we will present an overview of dataflow-based novel AI accelerators from SambaNova, Cerebras, Graphcore, and Groq. We evaluate these novel accelerators with diverse science workloads and compare the performance on GPUs. We discuss key insights, challenges, and opportunities in using these novel AI accelerators.

  Speaker Bio: Venkatram Vishwanath is a computer scientist at Argonne National Laboratory. He is the Data Science Team Lead at the Argonne leadership computing facility (ALCF). His current focus is on algorithms, system software, and workflows to facilitate data-centric applications on supercomputing systems. His interests include scientific applications, supercomputing architectures, parallel algorithms and runtimes, scalable analytics and collaborative workspaces. He has received best papers awards at venues including HPDC and LDAV, and a Gordon Bell finalist. Vishwanath received his Ph.D. in computer science from the University of Illinois at Chicago in 2009. His research Interests are: Scientific data analysis and visualization, Parallel I/O and I/O middleware, Large-scale computing systems and other exotic architectures (Blue Gene, Cray, multi-core systems, GPUs and other accelerators), and High-speed interconnects (InfiniBand, high-speed Ethernet, optical), data movement and transfer protocols, and (v) collaboration workspaces.