Design Automation of Approximate Circuits With Runtime Reconfigurable Accuracy
Leveraging the inherent error tolerance of a vast number of application domains that are rapidly growing, approximate computing arises as a design alternative to improve the efficiency of our computing systems by trading accuracy for energy savings. However, the requirement for computational accurac...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9039585/ |
| _version_ | 1818857156353458176 |
|---|---|
| author | Georgios Zervakis Hussam Amrouch Jorg Henkel |
| author_facet | Georgios Zervakis Hussam Amrouch Jorg Henkel |
| author_sort | Georgios Zervakis |
| collection | DOAJ |
| description | Leveraging the inherent error tolerance of a vast number of application domains that are rapidly growing, approximate computing arises as a design alternative to improve the efficiency of our computing systems by trading accuracy for energy savings. However, the requirement for computational accuracy is not fixed. Controlling the applied level of approximation dynamically at runtime is a key to effectively optimize energy, while still containing and bounding the induced errors at runtime. In this paper, we propose and implement an automatic and circuit independent design framework that generates approximate circuits with dynamically reconfigurable accuracy at runtime. The generated circuits feature varying accuracy levels, supporting also accurate execution. Extensive experimental evaluation, using industry strength flow and circuits, demonstrates that our generated approximate circuits improve the energy by up to 41% for 2% error bound and by 17.5% on average under a pessimistic scenario that assumes full accuracy requirement in the 33% of the runtime. To demonstrate further the efficiency of our framework, we considered two state-of-the-art technology libraries which are a 7nm conventional FinFET and an emerging technology that boosts performance at a high cost of increased dynamic power. |
| first_indexed | 2024-12-19T08:35:55Z |
| format | Article |
| id | doaj.art-aa0809b59d83434eb041244d113210a9 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-19T08:35:55Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-aa0809b59d83434eb041244d113210a92022-12-21T20:29:03ZengIEEEIEEE Access2169-35362020-01-018535225353810.1109/ACCESS.2020.29813959039585Design Automation of Approximate Circuits With Runtime Reconfigurable AccuracyGeorgios Zervakis0https://orcid.org/0000-0001-8110-7122Hussam Amrouch1https://orcid.org/0000-0002-5649-3102Jorg Henkel2https://orcid.org/0000-0001-9602-2922Department of Computer Science, Chair for Embedded Systems, Karlsruhe Institute of Technology, Karlsruhe, GermanyDepartment of Computer Science, Chair for Embedded Systems, Karlsruhe Institute of Technology, Karlsruhe, GermanyDepartment of Computer Science, Chair for Embedded Systems, Karlsruhe Institute of Technology, Karlsruhe, GermanyLeveraging the inherent error tolerance of a vast number of application domains that are rapidly growing, approximate computing arises as a design alternative to improve the efficiency of our computing systems by trading accuracy for energy savings. However, the requirement for computational accuracy is not fixed. Controlling the applied level of approximation dynamically at runtime is a key to effectively optimize energy, while still containing and bounding the induced errors at runtime. In this paper, we propose and implement an automatic and circuit independent design framework that generates approximate circuits with dynamically reconfigurable accuracy at runtime. The generated circuits feature varying accuracy levels, supporting also accurate execution. Extensive experimental evaluation, using industry strength flow and circuits, demonstrates that our generated approximate circuits improve the energy by up to 41% for 2% error bound and by 17.5% on average under a pessimistic scenario that assumes full accuracy requirement in the 33% of the runtime. To demonstrate further the efficiency of our framework, we considered two state-of-the-art technology libraries which are a 7nm conventional FinFET and an emerging technology that boosts performance at a high cost of increased dynamic power.https://ieeexplore.ieee.org/document/9039585/Approximate computingapproximate design automationdynamically reconfigurable accuracylow power |
| spellingShingle | Georgios Zervakis Hussam Amrouch Jorg Henkel Design Automation of Approximate Circuits With Runtime Reconfigurable Accuracy IEEE Access Approximate computing approximate design automation dynamically reconfigurable accuracy low power |
| title | Design Automation of Approximate Circuits With Runtime Reconfigurable Accuracy |
| title_full | Design Automation of Approximate Circuits With Runtime Reconfigurable Accuracy |
| title_fullStr | Design Automation of Approximate Circuits With Runtime Reconfigurable Accuracy |
| title_full_unstemmed | Design Automation of Approximate Circuits With Runtime Reconfigurable Accuracy |
| title_short | Design Automation of Approximate Circuits With Runtime Reconfigurable Accuracy |
| title_sort | design automation of approximate circuits with runtime reconfigurable accuracy |
| topic | Approximate computing approximate design automation dynamically reconfigurable accuracy low power |
| url | https://ieeexplore.ieee.org/document/9039585/ |
| work_keys_str_mv | AT georgioszervakis designautomationofapproximatecircuitswithruntimereconfigurableaccuracy AT hussamamrouch designautomationofapproximatecircuitswithruntimereconfigurableaccuracy AT jorghenkel designautomationofapproximatecircuitswithruntimereconfigurableaccuracy |