Managing the Performance/Error Tradeoff of Floating-point Intensive Applications
نویسندگان
چکیده
منابع مشابه
Performance Tradeoff Spectrum of Integer and Floating Point Applications Kernels on Various GPUs
Floating point precision and performance and the ratio of floating point units to integer processing elements on a graphics processing unit accelerator all continue to present complex tradeoffs for optimising core utilisation on modern devices. We investigate various hybrid CPU and GPU combinations using a range of different GPU models occupying different points in this tradeoff space. We analy...
متن کاملConverting floating - point applications to fixed - point
Without knowledge of the issues, the right tools, and a well-thought out development methodology, floating-point applications can in fact suffer the same problems as fixed-point equivalents. The reduced precision and range of fixed-point numbers means that fixed-point applications encounter stability problems more easily than floating-point applications, but both types of applications encounter...
متن کاملProduction Floating Point Applications on FPGAs
While FPGAs have only one fifth the raw floating point capability of GPUs, other attributes allow them to be surprisingly competitive with respect to a number of critical floating point intensive applications. In the first part we review these FPGA attributes. The bulk of this extended abstract then provides an overview of efficient FPGA implementations of Molecular Dynamics, Molecular Docking,...
متن کاملIntensive Applications in Self-Managing Computer Systems
rapid advances in processing power, network bandwidth, and storage capacity, computer are increasingly becoming extremely complex. Consequently, it becomes expensive and for human beings to manually manage complex computer systems. This problem can be ly tackled by self-managing computer systems, which are intended to meet high nce requirements in a dynamic computing environment. In this paper,...
متن کاملTradeoff of FPGA Design of a Floating-point Library for Arithmetic Operators
Most of the engineering and scientific applications involve the implementation of complex algorithms that are based on arithmetic operators [1]. The fixed-point arithmetic allows the computations to be performed with a high precision according to the bitwidth representation. However, many applications require to work not only with a high precision, but also with a suitable format in order to re...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: ACM Transactions on Embedded Computing Systems
سال: 2017
ISSN: 1539-9087,1558-3465
DOI: 10.1145/3126519