Many-core acceleration in biomedical applications

David Kaeli

Professor of Electrical and Computer Engineering, Northeastern University

Abstract

1. Graphic Processing Units (GPUs) have been shown to provide significant opportunities for overcoming computational barriers in a number of critical applications domains.  GPUs exploit massive data-level and thread-level parallelism to obtain impressive speedups. With the introduction of user-friendly programming frameworks such as CUDA and OpenCL, GPUs have been used to accelerate a range of challenging computational problems.
   

2. In this talk we will focus on our recent work accelerating biomedical applications.  We will present some of the key challenges in these applications, and show how GPUs have been used to enable near real-time performance in a number of critical biomedical domains.  We will also discuss how we achieve some of these impressive speedups through improved memory mapping, library specialization, and multi-GPU acceleration.
   

About the speaker

1. David Kaeli received a BS and PhD in Electrical Engineering from Rutgers University, and an MS in Computer Engineering from Syracuse University.  He is presently a Full Processor on the ECE faculty at Northeastern University, Boston, MA where he directs the Northeastern University Computer Architecture Research Laboratory (NUCAR).  Prior to joining Northeastern in 1993, Kaeli spent 12 years at IBM, the last 7 at T.J. Watson Research Center, Yorktown Heights, NY.
   

2. Dr. Kaeli has published over 180 critically reviewed publications, 6 books, and 8 patents.  He regularly provides tutorials on the subject of Profiling and Instrumentation, and has organized workshops on GPGPUs, security, and binary translation. His research spans a wide range of areas including VLSI design to back-end complers and database systems. He is an Associate Editor of IEEE Computer Architecture Letters and the Journal of Instruction Level Parallelism, and the Chair of the IEEE Technical Committee on Microprocessors and Microprogramming. He also serves as a member of CRA's Computing Consortium Council.