Survey of deployment locations and underlying hardware architectures for contemporary deep neural networks
This article overviews the emerging use of deep neural networks in data analytics and explores which type of underlying hardware and architectural approach is best used in various deployment locations when implementing deep neural networks. The locations which are discussed are in the cloud, fog, an...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Hindawi - SAGE Publishing
2019-08-01
|
| Series: | International Journal of Distributed Sensor Networks |
| Online Access: | https://doi.org/10.1177/1550147719868669 |
| _version_ | 1797711479020978176 |
|---|---|
| author | Miloš Kotlar Dragan Bojić Marija Punt Veljko Milutinović |
| author_facet | Miloš Kotlar Dragan Bojić Marija Punt Veljko Milutinović |
| author_sort | Miloš Kotlar |
| collection | DOAJ |
| description | This article overviews the emerging use of deep neural networks in data analytics and explores which type of underlying hardware and architectural approach is best used in various deployment locations when implementing deep neural networks. The locations which are discussed are in the cloud, fog, and dew computing (dew computing is performed by end devices). Covered architectural approaches include multicore processors (central processing unit), manycore processors (graphics processing unit), field programmable gate arrays, and application-specific integrated circuits. The proposed classification in this article divides the existing solutions into 12 different categories, organized in two dimensions. The proposed classification allows a comparison of existing architectures, which are predominantly cloud-based, and anticipated future architectures, which are expected to be hybrid cloud-fog-dew architectures for applications in Internet of Things and Wireless Sensor Networks. Researchers interested in studying trade-offs among data processing bandwidth, data processing latency, and processing power consumption would benefit from the classification made in this article. |
| first_indexed | 2024-03-12T07:07:42Z |
| format | Article |
| id | doaj.art-f8f88b3a9aa74c8ba37483d7caedba63 |
| institution | Directory Open Access Journal |
| issn | 1550-1477 |
| language | English |
| last_indexed | 2024-03-12T07:07:42Z |
| publishDate | 2019-08-01 |
| publisher | Hindawi - SAGE Publishing |
| record_format | Article |
| series | International Journal of Distributed Sensor Networks |
| spelling | doaj.art-f8f88b3a9aa74c8ba37483d7caedba632023-09-02T23:20:49ZengHindawi - SAGE PublishingInternational Journal of Distributed Sensor Networks1550-14772019-08-011510.1177/1550147719868669Survey of deployment locations and underlying hardware architectures for contemporary deep neural networksMiloš Kotlar0Dragan Bojić1Marija Punt2Veljko Milutinović3School of Electrical Engineering, University of Belgrade, Belgrade, SerbiaSchool of Electrical Engineering, University of Belgrade, Belgrade, SerbiaSchool of Electrical Engineering, University of Belgrade, Belgrade, SerbiaDepartment of Computer Science, Indiana University Bloomington, Bloomington, IN, USAThis article overviews the emerging use of deep neural networks in data analytics and explores which type of underlying hardware and architectural approach is best used in various deployment locations when implementing deep neural networks. The locations which are discussed are in the cloud, fog, and dew computing (dew computing is performed by end devices). Covered architectural approaches include multicore processors (central processing unit), manycore processors (graphics processing unit), field programmable gate arrays, and application-specific integrated circuits. The proposed classification in this article divides the existing solutions into 12 different categories, organized in two dimensions. The proposed classification allows a comparison of existing architectures, which are predominantly cloud-based, and anticipated future architectures, which are expected to be hybrid cloud-fog-dew architectures for applications in Internet of Things and Wireless Sensor Networks. Researchers interested in studying trade-offs among data processing bandwidth, data processing latency, and processing power consumption would benefit from the classification made in this article.https://doi.org/10.1177/1550147719868669 |
| spellingShingle | Miloš Kotlar Dragan Bojić Marija Punt Veljko Milutinović Survey of deployment locations and underlying hardware architectures for contemporary deep neural networks International Journal of Distributed Sensor Networks |
| title | Survey of deployment locations and underlying hardware architectures for contemporary deep neural networks |
| title_full | Survey of deployment locations and underlying hardware architectures for contemporary deep neural networks |
| title_fullStr | Survey of deployment locations and underlying hardware architectures for contemporary deep neural networks |
| title_full_unstemmed | Survey of deployment locations and underlying hardware architectures for contemporary deep neural networks |
| title_short | Survey of deployment locations and underlying hardware architectures for contemporary deep neural networks |
| title_sort | survey of deployment locations and underlying hardware architectures for contemporary deep neural networks |
| url | https://doi.org/10.1177/1550147719868669 |
| work_keys_str_mv | AT miloskotlar surveyofdeploymentlocationsandunderlyinghardwarearchitecturesforcontemporarydeepneuralnetworks AT draganbojic surveyofdeploymentlocationsandunderlyinghardwarearchitecturesforcontemporarydeepneuralnetworks AT marijapunt surveyofdeploymentlocationsandunderlyinghardwarearchitecturesforcontemporarydeepneuralnetworks AT veljkomilutinovic surveyofdeploymentlocationsandunderlyinghardwarearchitecturesforcontemporarydeepneuralnetworks |