A Proposed Radix- and Word-length-independent Standard for Floating-point Arithmetic

Cody, W. J.; Coonen, Jerome T.; Hanson, Kenton; Hough, David; Kahan, W.; Karpinski, Richard; Palmer, J.; Ris, Frederic N.; Stevenson, D. P.

doi:10.1109/mm.1984.291224

Cited by 35 publications

(14 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another example is a fluid simulation that is coupled with a solids structure code, as is done in some industrial process modeling [8]. We also examine performance for large arrays of IEEE 754 standard doubles [15] (64-bit floats) which routinely occur in scientific computing and typically dominate the total number of bytes sent between procedures and functions. We varied the size of the double arrays from 10 to 1,000,000.…”

Section: Schema-specific Parsersmentioning

confidence: 99%

Investigating the limits of SOAP performance for scientific computing

Chiu

Govindaraju

Bramley

Proceedings 11th IEEE International Symposium on High Performance Distributed Computing

186

137

View full text Add to dashboard Cite

The growing synergy between Web Services and Gridbased technologies [7] will potentially enable profound, dynamic interactions between scientific applications dispersed in geographic, institutional, and conceptual space. Such deep interoperability requires the simplicity, robustness, and extensibility for which SOAP [4,3] was conceived, thus making it a natural lingua franca. Concomitant with these advantages, however, is a degree of inefficiency that may limit the applicability of SOAP to some situations. In this paper, we investigate the limitations of SOAP for high-performance scientific computing. We analyze the processing of SOAP messages, and identify the issues of each stage. We present a high-performance SOAP implementation and a schema-specific parser based on the results of our investigation. After our SOAP optimizations are implemented, the most significant bottleneck is ASCII/double conversion. Instead of handling this using extensions to SOAP, we recommend a multiprotocol approach that uses SOAP to negotiate faster binary protocols between messaging participants.

show abstract

Section: Schema-specific Parsersmentioning

confidence: 99%

Investigating the limits of SOAP performance for scientific computing

Chiu

Govindaraju

Bramley

Proceedings 11th IEEE International Symposium on High Performance Distributed Computing

186

137

View full text Add to dashboard Cite

show abstract

“…We considered some alternatives, such as adding IEEE floating point not-a-number (NAN) style values [7] would solve the iteration problem, but would introduce a larger problem because the number of bits required to represent bi-valued enumerations like bit and boolean would double. While not a problem for scalars, this would certainly be problematic for arrays, especially when used to represent numeric data types like ubyte and int.…”

Section: Iterationmentioning

confidence: 99%

Kava

Bacon¹

2001

Proceedings of the 2001 Joint ACM-ISCOPE Conference on Java Grande

View full text Add to dashboard Cite

ObJect-oriented programming languages have always distinguished between "primitive" and "user-defined" data types, and in the case of languages like C++ and Java, the primitives are not even treated as objects, further fragmenting the programming model. The distinction is especially problematic when a particular programrning community requires primitive-level support for a new data type, as for complex, intervals, fixed-pointed numbers, and so on.We present Kava, a design for a backward-compatible version of Java that solves the problem of programmable lightweight objects in a much more aggressive and uniform manner than previous proposals. In Kava, there are no primitive types; instead, objectoriented programming is provided down to the level of single bits, and types such as int can be explicitly programmed within the language. While the language maintains a uniform object reference semantics, efficiency is obtained by making heavy use of unboxing and semantic expansion.We describe Kava as a dialect of the Java language, show how it can be used to define various primitive types, describe how it can be translated into Java, and compare it to other approaches to lightweight objects.

show abstract

“…Exponent reduction by means of the factors described above reduces the absolute value of the decimal exponent by 9, 13, or 14 per step, so the last step (to reduce the remaining decimal exponent to zero) may have to use a different factor, chosen from a small table of residual factors of the same general form. For example, to reduce a residual decimal exponent of ϩ5, use the factor 5 5 Note that, because of the choice of decimal arithmetic for reducing negative exponents, all calculations described above are exact if we allow the bignum fraction to grow by up to one bigdigit per multiplication. The maximum number of multiplications is determined by the valid exponent range, and is therefore bounded by the format of the binary floating-point number (whether input or output).…”

Section: Conversion Between Decimal and Binary Floating-pointmentioning

confidence: 99%

“…This work was conducted through the IEEE Floating-Point Working Group P754. IBM's first "straw-man" architecture proposal for implementing the IEEE Floating-Point draft standard [1][2][3][4][5] in the then-current S/370* architecture dates from 1982. In 1985, ANSI/IEEE Standard 754-1985 [6] was approved.…”

Section: Introductionmentioning

confidence: 99%