Issue No.01 - January-June (2007 vol.6)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/L-CA.2007.1
Some processors designed for consumer applications, such as Graphics Processing Units (GPUs) and the CELL processor, promise outstanding floating-point performance for scientific applications at commodity prices. However, IEEE single precision is the most precise floating-point data type these processors directly support in hardware. Pairs of native floatingpoint numbers can be used to represent a base result and a residual term to increase accuracy, but the resulting order of magnitude slowdown dramatically reduces the price/performance advantage of these systems. By adding a few simple microarchitectural features, acceptable accuracy can be obtained with relatively little performance penalty. To reduce the cost of native-pair arithmetic, a residual register is used to hold information that would normally have been discarded after each floating-point computation. The residual register dramatically simplifies the code, providing both lower latency and better instruction-level parallelism.
Cost/performance, High-Speed Arithmetic, Arithmetic and Logic Structures, Hardware, Hardware/software interfaces, General, Computer Systems Organization, Micro-architecture implementation considerations, Processor Architectures, Computer Systems Organization, Multiple precision arithmetic, General, Numerical Analysis, Mathematics of Computing, Graphics processors, Hardware Architecture, Computer Graphics, Computing Methodologies
William R. Dieter, Henry G. Dietz, "Low-Cost Microarchitectural Support for Improved Floating-Point Accuracy", IEEE Computer Architecture Letters, vol.6, no. 1, pp. 13-16, January-June 2007, doi:10.1109/L-CA.2007.1