2010 International Conference on Field Programmable Logic and Applications 2010
DOI: 10.1109/fpl.2010.15
|View full text |Cite
|
Sign up to set email alerts
|

Synthesis of Floating-Point Addition Clusters on FPGAs Using Carry-Save Arithmetic

Abstract: Abstract-A new method to synthesize clusters of floating-point addition operations on FPGAs is presented. Similar to Altera's floating-point data path compiler, it performs normalization once, at the output of the cluster operation. All significands in the clustered operation are denormalized in parallel with respect to the largest exponent: a fixed-point compressor tree then sums the aligned significands, followed by normalization and rounding. Compared to Altera's floating-point datapath compiler, our method… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
4
0

Year Published

2012
2012
2014
2014

Publication Types

Select...
2
2

Relationship

1
3

Authors

Journals

citations
Cited by 4 publications
(4 citation statements)
references
References 8 publications
0
4
0
Order By: Relevance
“…First, a hardware summation operator can take various forms: a multi-input adder (the operator handles a limited number of addends at once) [Tenca 2009;Verma et al 2010], an accumulator (operands are treated serially and their number can be very large) [Kulisch 2002;Luo and Martonosi 2000], a composite of the two operators [Morris and Prasanna 2005;Zhuo and Prasanna 2008], and other hardware approaches [Kapre and DeHon 2007;Tai et al 2012].…”
Section: Hardware Summationmentioning
confidence: 99%
“…First, a hardware summation operator can take various forms: a multi-input adder (the operator handles a limited number of addends at once) [Tenca 2009;Verma et al 2010], an accumulator (operands are treated serially and their number can be very large) [Kulisch 2002;Luo and Martonosi 2000], a composite of the two operators [Morris and Prasanna 2005;Zhuo and Prasanna 2008], and other hardware approaches [Kapre and DeHon 2007;Tai et al 2012].…”
Section: Hardware Summationmentioning
confidence: 99%
“…Let e q be the quantization error defined as the difference between the data x and x Q . The data x can be modeled as the sum of the quantization process x Q and the quantization error e q (Verma, 2010).…”
Section: Floating-point and Fixed-point Codingmentioning
confidence: 99%
“…However, it has focused on the general synthesis of CS structures, not their selective use to accelerate floating-point operations. Other approaches use CS arithmetic internally to individual operations, but not between them [12,22].…”
Section: Related Workmentioning
confidence: 99%