2008
DOI: 10.1007/978-3-540-85373-2_8
|View full text |Cite
|
Sign up to set email alerts
|

The Reduceron: Widening the von Neumann Bottleneck for Graph Reduction Using an FPGA

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
16
0

Year Published

2008
2008
2023
2023

Publication Types

Select...
4
2
1

Relationship

2
5

Authors

Journals

citations
Cited by 15 publications
(16 citation statements)
references
References 6 publications
0
16
0
Order By: Relevance
“…stack size), even though there are successful attempts to create large architectures which could handle this to some degree, like the functional neurons of SyNAPSE architecture by IBM [2]. FPGA-based functional architectures are also attempted, such as the Reduceron [120] and functional design of reconfigurable architectures has been thoroughly explored [29]. Recombination of complete kernels to compose more complex applications by chaining inputs and outputs in a configurable fashion within a processor pipeline (Function Level Processor) has been also attempted with great results, but it does not apply functional programming theory [155].…”
Section: The Von Neumann Bottleneckmentioning
confidence: 99%
“…stack size), even though there are successful attempts to create large architectures which could handle this to some degree, like the functional neurons of SyNAPSE architecture by IBM [2]. FPGA-based functional architectures are also attempted, such as the Reduceron [120] and functional design of reconfigurable architectures has been thoroughly explored [29]. Recombination of complete kernels to compose more complex applications by chaining inputs and outputs in a configurable fashion within a processor pipeline (Function Level Processor) has been also attempted with great results, but it does not apply functional programming theory [155].…”
Section: The Von Neumann Bottleneckmentioning
confidence: 99%
“…This approach (Naylor & Runciman, 2008) is more efficient in some cases, but it cannot be expressed as a core-language transformation.…”
Section: Bounded Template Instantiationmentioning
confidence: 99%
“…Compared to the Reduceron presented in Naylor & Runciman (2008), the implementation described in this paper reduces the number of clock-cycles required to run the benchmark programs by an average factor of 6.4. As the previous implementation clocks at 111 MHz, and the new one at 96 MHz on the same FPGA, the raw speedup factor is 5.5.…”
Section: Comparison With the 2007 Reduceronmentioning
confidence: 99%
See 1 more Smart Citation
“…SmallCheck is more effective on strict properties and Lazy SmallCheck wins for lazy ones. Of these examples, ListSet is the set implementation using ordered lists (along with the insertion property) given earlier, Countdown is a solver for a popular numbers game (along with a lemma and a refinement theorem) taken from (Hutton 2002), SumPuz is a cryptarithmetic solver (with a soundness property) from (Claessen et al 2002), Circuits is part of a library from the Reduceron (Naylor and Runciman 2008), and Catch is a specification (with a soundness property) for part of the Catch tool (Mitchell and Runciman 2007).…”
Section: Other Examplesmentioning
confidence: 99%