Floating Point Arithmetic, GPUs and Biomedical Imaging Applications

Abstract

GPU and multicore architectures are often used to accelerate scientific applications that rely on floating point computation. Current techniques for verifying the correctness of these implementations are quite primitive. The same algorithm run on different architectures will produce different results. This is due to the nature of floating point arithmetic and not the implementation of individual floating point operations. Programmers who are familiar with this expect their floating point code to produce different results on different architectures, and as a result do not test their code thoroughly. This talk will consider several case studies to analyze the sources of differences in CPU and GPU code. A brief survey of tools available for addressing these issues and current research directions in identifying bugs in GPU and multicore implementations will be presented.

In the second part of the talk, I will talk about some projects we have to accelerate scientific applications on GPUs, including Monte Carlo Simulation for Diffuse Optical Tomography, 3D Computed Tomography Reconstruction, and lung tumor tracking.

Bio Professor Miriam Leeser received the BS degree in Electrical Engineering from Cornell University and the Diploma and PhD in Computer Science from Cambridge University, England. In 1992, she received a National Science Foundation CAREER award to conduct research into floating point arithmetic. She has been on the faculty of Northeastern since 1996, where she is head of the Reconfigurable and GPU Computing Laboratory and a member of the computer engineering research group and the Center for Communications and Digital Signal Processing. She conducts research into accelerating image and signal processing applications with nontraditional computer architectures, including FPGAs and GPUs.