Implementation Of Pipelining for Modified Fused Multiply Add Architecture

Duvvuru Praveen Kumar*, M. Bharathi**
* M.Tech Student, VLSI, Department of Electronics and Communication Engineering, Sree Vidyanikethan Engineering College, Tirupathi.
** Assistant Professor, Department of Electronics and Communication Engineering, Sree Vidyanikethan Engineering College, Tirupathi.
Periodicity:September - November'2014
DOI : https://doi.org/10.26634/jcir.2.4.3219

Abstract

Fused Multiply Add (FMA) units generally reduce delay in the overall circuit and also make it efficient in terms of energy in arithmetic operations. In this paper, the authors present an effective implementation of MFMA (Modified FMA) using pipelining by decreasing the delay compared to the parallel processing of the module. This MFMA uses two desperate Massif which are connected using pipelining. The final operation performed is A*B+C*D+F*G+H*I, parallel processing just implements unto A*B+C*D. The clock limiting stage for both these operations is involved in normalization stage and rounding stage. This paper is related to floating point calculations. Floating point calculations involve, a standard format for representing floating point numbers. The standard format for representing floating point numbers is IEEE 754- 2008.This floating point representation is used here. In this paper the pipelining implementation is mainly related to speed, i.e., delay of the circuit. This paper can be designed using trilogy HDL or VHDL and is simulated and synthesized in XILINX ISE 10.1 of FPGA.

Keywords

FMA, MFMA, MAC, Cascaded FMA.

How to Cite this Article?

Kumar, D. P., and Bharathi, M. (2014). Implementation Of Pipelining for Modified Fused Multiply Add Architecture. i-manager’s Journal on Circuits and Systems, 2(4), 7-12. https://doi.org/10.26634/jcir.2.4.3219

References

[1]. Siberian Aspirin, Oren Legal, Martin Malarkey, (2014). "Modified Fused Multiply - Accumulate Chained Unit". 978- 1-4799-4132-2/14 @2014 IEEE.
[2]. Sam eh Algal and Mark Horowitz ( 2011). "Energy Efficient Floating Point Unit Design". IEEE Transactions On Computers, Vol. 60, No. 7, pp. 913-922.
[3]. www.wikipedia.com.
[4]. Eng. Walaa Abd El Aziz Ibrahim (2011). "Binary Floating Point Fused Multiply Add Unit", Thesis submitted to Cairo University.
[5]. Tom M. Bruises (2011). “Design of a Fused Multiply-Add Floating-Point and Integer Apathy,” Master Thesis.
[6]. Afar M. ALGOL (2008). "Optimization Effects on Mode ling and Synthesis of a Conventional Floating Point Fused Multiply-Add Arithmetic Unit Using CAD Tools". Journal of disciplinarian & multidisciplinary research, Vol. 2, No. 1.
[7]. Tomas Lang and Wavier D. Beleaguer (2004). “Floating-point multiply-add fused with reduced latency”, IEEE Transactions on Computers, Vol. 53, No 8, pp. 988–1003.
[8]. R.K. Montoya, E. Hoboken and S.L. Runyon (1990). “Design of the IBMRISC System/6000 floating-point execution unit,” IBM Journal of Research & Development, Vol. 34, pp. 59-70.
[9]. B. Jeevan,S. Narender, Dr C.V. Krishna Reddy, Dr K. Sivani, (2013). “A High Speed Binary Floating Point Multiplier Using Dadda Algorithm”, IEEE Transaction, pp.455-460.
If you have access to this article please login to view the article or kindly login to purchase the article

Purchase Instant Access

Single Article

North Americas,UK,
Middle East,Europe
India Rest of world
USD EUR INR USD-ROW
Pdf 35 35 200 20
Online 35 35 200 15
Pdf & Online 35 35 400 25

Options for accessing this content:
  • If you would like institutional access to this content, please recommend the title to your librarian.
    Library Recommendation Form
  • If you already have i-manager's user account: Login above and proceed to purchase the article.
  • New Users: Please register, then proceed to purchase the article.