Academia.eduAcademia.edu

Figure 3 – uploaded by Dimitrios Soudris

See full PDFdownloadDownload figure
somewhat greater than that of the Xilinx architecture for benchmarks after b14. Once again, this is due to the tool limitations that lead to an increased number of LUTs. Still, it can be seen that the relative increase in power consump- tion per benchmark is smaller than the relative increase in number of LUTs (35% and 25% respectively in the case of benchmark b_20) which confirms the efficiency of the employed circuit-level techniques. In order to improve the power efficiency of the proposed system, the LUT-mapping process of E2FMT and DRUID will have to be improved.

Table 3 somewhat greater than that of the Xilinx architecture for benchmarks after b14. Once again, this is due to the tool limitations that lead to an increased number of LUTs. Still, it can be seen that the relative increase in power consump- tion per benchmark is smaller than the relative increase in number of LUTs (35% and 25% respectively in the case of benchmark b_20) which confirms the efficiency of the employed circuit-level techniques. In order to improve the power efficiency of the proposed system, the LUT-mapping process of E2FMT and DRUID will have to be improved.

Related Figures (20)

all the BLEs [8], results in an almost linear dependency with the number of LUT inputs, and the cluster size, considering the formula:
all the BLEs [8], results in an almost linear dependency with the number of LUT inputs, and the cluster size, considering the formula:
the entire system of tools and platform. Conclusions are further discussed in Sect. 5.
the entire system of tools and platform. Conclusions are further discussed in Sect. 5.
Table 1 Power gains achieved by clock gating.
Table 1 Power gains achieved by clock gating.
Fig.4 Impact of SB type and length on energy-delay product.
Fig.4 Impact of SB type and length on energy-delay product.
Fig.6 Performance exploration results. Fig.5 | Energy consumption exploration results.
Fig.6 Performance exploration results. Fig.5 | Energy consumption exploration results.
Fig.7 Area exploration results.
Fig.7 Area exploration results.
Fig.9 |The configuration architecture.
Fig.9 |The configuration architecture.
Fig.11 The proposed design framework.
Fig.11 The proposed design framework.
Fig.12 | DAGGER flowchart.
Fig.12 | DAGGER flowchart.
Fig.13 Partial reconfiguration flowchart.
Fig.13 Partial reconfiguration flowchart.
Table 2 Qualitative comparison among tool frameworks. tations of the proposed framework are that it does not cur- rently support back-annotation, but no other academic tool frameworks do either. It is evident that the proposed tool framework is the most complete academic tool framework, and is at least in terms of provided features comparable with commercial tools. It contains the only known academic implementation of a configuration bitstream generation tool. Additionally, the remote access to GUI feature allows the user to run the framework without even having the tools installed in his/her own computer.
Table 2 Qualitative comparison among tool frameworks. tations of the proposed framework are that it does not cur- rently support back-annotation, but no other academic tool frameworks do either. It is evident that the proposed tool framework is the most complete academic tool framework, and is at least in terms of provided features comparable with commercial tools. It contains the only known academic implementation of a configuration bitstream generation tool. Additionally, the remote access to GUI feature allows the user to run the framework without even having the tools installed in his/her own computer.
Fig.16 | Power consumption comparison. due to the fact that the E2FMT tool libraries do not support many basic modules that had to be added by DRUID de- scribed at gate level, which leads to larger netlists and there- fore greater number of LUTs. This can only be efficiently remedied if E2FMT is drastically modified.
Fig.16 | Power consumption comparison. due to the fact that the E2FMT tool libraries do not support many basic modules that had to be added by DRUID de- scribed at gate level, which leads to larger netlists and there- fore greater number of LUTs. This can only be efficiently remedied if E2FMT is drastically modified.
Fig.15 = Maximum frequency comparison.
Fig.15 = Maximum frequency comparison.
Fig.14 LUT mapping comparison.
Fig.14 LUT mapping comparison.
Fig.17 | Low-swing power savings.
Fig.17 | Low-swing power savings.
Haroula Pournara received the B.Sc. degree in Physics and the M.Sc. degree in electronics engineering from the Aristotle University of Thessa- loniki, Greece, in 2001 and 2004, respectively.
Haroula Pournara received the B.Sc. degree in Physics and the M.Sc. degree in electronics engineering from the Aristotle University of Thessa- loniki, Greece, in 2001 and 2004, respectively.
Related topics:
Information SystemsLawField Programmable Gate ArrayDigital LogicE-SystemsCircuit DesignElectrical and Electronic EngineeringConfigurable Logic Block (CLB)FPGA ArchitecturePower Added Efficiency
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved