The Parallel Evaluation of Arithmetic Expressions Without Division
نویسندگان
چکیده
As computers become capable of executing more arithmetic operations simultaneously, the question of compiling for such machines becomes more important. In this correspondence we consider arbitrary arithmetic expressions of n distinct variables with operations restricted to addition, subtraction, and multiplication. We first construct a scheme whereby any such expression can be evaluated in at most 3 log2 n + O(1) steps if sufficiently many processors are available. We then improve this result and reduce 3 log2 n to 2.465 log2 n. Finally, we deduce some results that apply when a fixed number of processors is available. CommentsOnly the Abstract is given here. The full paper appeared as [1] and was reprinted in [3, pages30–35]. For a more general result which allows division, and requires less processors, see [2]. References[1] R. P. Brent, D. J. Kuck and K. Maruyama, “The parallel evaluation of arithmetic expressions without division”,IEEE Transactions on Computers C-22 (1973), 532–534. MR 50#11843. rpb015.[2] R. P. Brent, “The parallel evaluation of general arithmetic expressions”, Journal of the ACM 21 (1974),201–206. CR 15#27055, MR 58#31996, Zbl 276.68010. rpb022.[3] R. P. Brent, Topics in computational complexity and the analysis of algorithms, Report TR-CS-80-14, DCS,ANU, October 1980, 375 pp. (D. Sc. thesis). rpb062. (R. P. Brent) Computer Centre, Australian National University, Canberra, Australia (D. J. Kuck and K. Maruyama) Department of Computer Science, University of Illinois, Urbana,Illinois 61801 1991 Mathematics Subject Classification. Primary 68Q22; Secondary 65Y05, 65Y20, 68M07, 68Q25.
منابع مشابه
Non radial model of dynamic DEA with the parallel network structure
In this article, Non radial method of dynamic DEA with the parallel network structure is presented and is used for calculation of relative efficiency measures when inputs and outputs do not change equally. In this model, DMU divisions under evaluation have been put together in parallel. But its dynamic structure is assumed in series. Since in real applications there are undesirable inputs an...
متن کاملLow Complexity Converter for the Moduli Set {2^n+1,2^n-1,2^n} in Two-Part Residue Number System
Residue Number System is a kind of numerical systems that uses the remainder of division in several different moduli. Conversion of a number to smaller ones and carrying out parallel calculations on these numbers will increase the speed of the arithmetic operations in this system. However, the main factor that affects performance of system is hardware complexity of reverse converter. Reverse co...
متن کاملA Note on the First Geometric-Arithmetic Index of Hexagonal Systems and Phenylenes
The first geometric-arithmetic index was introduced in the chemical theory as the summation of 2 du dv /(du dv ) overall edges of the graph, where du stand for the degree of the vertex u. In this paper we give the expressions for computing the first geometric-arithmetic index of hexagonal systems and phenylenes and present new method for describing hexagonal system by corresponding a simple g...
متن کاملA High-Speed Dual-Bit Parallel Adder based on Carbon Nanotube FET technology for use in arithmetic units
In this paper, a Dual-Bit Parallel Adder (DBPA) based on minority function using Carbon-Nanotube Field-Effect Transistor (CNFET) is proposed. The possibility of having several threshold voltage (Vt) levels by CNFETs leading to wide use of them in designing of digital circuits. The main goal of designing proposed DBPA is to reduce critical path delay in adder circuits. The proposed design positi...
متن کاملOverflow Detection in Residue Number System, Moduli Set {2n-1,2n,2n+1}
Residue Number System (RNS) is a non-weighted number system for integer number arithmetic, which is based on the residues of a number to a certain set of numbers called module set. The main characteristics and advantage of residue number system is reducing carry propagation in calculations. The elimination of carry propagation leads to the possibility of maximizing parallel processing and reduc...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
- IEEE Trans. Computers
دوره 22 شماره
صفحات -
تاریخ انتشار 1973