Issue No. 08 - Aug. (2012 vol. 61)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TC.2012.89
Abhilasha Aswal , International Institute of Information Technology, Bangalore and Infosys Limited, Bangalore
M. Ganesh Perumal , International Institute of Information Technology, Bangalore and Infosys Limited, Bangalore
G.N. Srinivasa Prasanna , International Institute of Information Technology, Bangalore
Financial transactions are specified in decimal arithmetic. Until the introduction of IEEE 754-2008, specialized software/hardware routines were used to perform these transactions but it incurred a penalty on performance. In this paper, we show that if binary arithmetic is used to emulate decimal operations, then arbitrary error sequences can be generated by carefully chosen sequences of transactions which can lead to monotonically increasing/decreasing capitalization errors. In addition, we describe methods for correctly performing basic decimal operations, such as addition, subtraction, multiplication, and division, on binary machines, which are not conformant with IEEE 754-2008 decimal floating point standard (ISO/IEC/IEEE 60559:2011), at high speed.
Floating-point arithmetic, IEEE 754 standards, computer arithmetic.
A. Aswal, M. G. Perumal and G. S. Prasanna, "On Basic Financial Decimal Operations on Binary Machines," in IEEE Transactions on Computers, vol. 61, no. , pp. 1084-1096, 2012.