Runtime DNN performance scaling through resource management on heterogeneous embedded platforms Article Swipe
DNN inference is increasingly being executed locally on embedded platforms, due to the clear advantages in latency, privacy and connectivity. Modern SoCs typically execute a combination of different and dynamic workloads concurrently, it is challenging to consistently meet latency/energy budgets because the local computing resources available to the DNN vary considerably. In this poster, we show how resource management can be applied to optimise the performance of DNN workloads by monitoring and tuning both software and hardware constantly at runtime. This work shows how dynamic DNNs trade-off accuracy with latency/energy/power on heterogeneous embedded CPU-GPU platform.
Related Topics
Concepts
Metadata
- Type
- article
- Language
- en
- https://eprints.soton.ac.uk/450052/1/Runtime_DNN_Performance_Scaling_through_Resource_Management_on_Heterogeneous_Embedded_Platforms.pdf
- OA Status
- green
- References
- 2
- Related Works
- 20
- OpenAlex ID
- https://openalex.org/W3181572714
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W3181572714Canonical identifier for this work in OpenAlex
- Title
-
Runtime DNN performance scaling through resource management on heterogeneous embedded platformsWork title
- Type
-
articleOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2021Year of publication
- Publication date
-
2021-06-10Full publication date if available
- Authors
-
Xun Lei, Bashir M. Al‐Hashimi, Jonathon Hare, Geoff V. MerrettList of authors in order
- PDF URL
-
https://eprints.soton.ac.uk/450052/1/Runtime_DNN_Performance_Scaling_through_Resource_Management_on_Heterogeneous_Embedded_Platforms.pdfDirect link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
greenOpen access status per OpenAlex
- OA URL
-
https://eprints.soton.ac.uk/450052/1/Runtime_DNN_Performance_Scaling_through_Resource_Management_on_Heterogeneous_Embedded_Platforms.pdfDirect OA link when available
- Concepts
-
Computer science, Latency (audio), Inference, Embedded system, Distributed computing, Software, Power management, Operating system, Power (physics), Artificial intelligence, Quantum mechanics, Telecommunications, PhysicsTop concepts (fields/topics) attached by OpenAlex
- Cited by
-
0Total citation count in OpenAlex
- References (count)
-
2Number of works referenced by this work
- Related works (count)
-
20Other works algorithmically related by OpenAlex
Full payload
| id | https://openalex.org/W3181572714 |
|---|---|
| doi | |
| ids.mag | 3181572714 |
| ids.openalex | https://openalex.org/W3181572714 |
| fwci | 0.0 |
| type | article |
| title | Runtime DNN performance scaling through resource management on heterogeneous embedded platforms |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| topics[0].id | https://openalex.org/T10502 |
| topics[0].field.id | https://openalex.org/fields/22 |
| topics[0].field.display_name | Engineering |
| topics[0].score | 0.9998999834060669 |
| topics[0].domain.id | https://openalex.org/domains/3 |
| topics[0].domain.display_name | Physical Sciences |
| topics[0].subfield.id | https://openalex.org/subfields/2208 |
| topics[0].subfield.display_name | Electrical and Electronic Engineering |
| topics[0].display_name | Advanced Memory and Neural Computing |
| topics[1].id | https://openalex.org/T10036 |
| topics[1].field.id | https://openalex.org/fields/17 |
| topics[1].field.display_name | Computer Science |
| topics[1].score | 0.9995999932289124 |
| topics[1].domain.id | https://openalex.org/domains/3 |
| topics[1].domain.display_name | Physical Sciences |
| topics[1].subfield.id | https://openalex.org/subfields/1707 |
| topics[1].subfield.display_name | Computer Vision and Pattern Recognition |
| topics[1].display_name | Advanced Neural Network Applications |
| topics[2].id | https://openalex.org/T12808 |
| topics[2].field.id | https://openalex.org/fields/22 |
| topics[2].field.display_name | Engineering |
| topics[2].score | 0.9977999925613403 |
| topics[2].domain.id | https://openalex.org/domains/3 |
| topics[2].domain.display_name | Physical Sciences |
| topics[2].subfield.id | https://openalex.org/subfields/2208 |
| topics[2].subfield.display_name | Electrical and Electronic Engineering |
| topics[2].display_name | Ferroelectric and Negative Capacitance Devices |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C41008148 |
| concepts[0].level | 0 |
| concepts[0].score | 0.8175351023674011 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q21198 |
| concepts[0].display_name | Computer science |
| concepts[1].id | https://openalex.org/C82876162 |
| concepts[1].level | 2 |
| concepts[1].score | 0.7684193849563599 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q17096504 |
| concepts[1].display_name | Latency (audio) |
| concepts[2].id | https://openalex.org/C2776214188 |
| concepts[2].level | 2 |
| concepts[2].score | 0.5852155089378357 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q408386 |
| concepts[2].display_name | Inference |
| concepts[3].id | https://openalex.org/C149635348 |
| concepts[3].level | 1 |
| concepts[3].score | 0.5152907371520996 |
| concepts[3].wikidata | https://www.wikidata.org/wiki/Q193040 |
| concepts[3].display_name | Embedded system |
| concepts[4].id | https://openalex.org/C120314980 |
| concepts[4].level | 1 |
| concepts[4].score | 0.48680922389030457 |
| concepts[4].wikidata | https://www.wikidata.org/wiki/Q180634 |
| concepts[4].display_name | Distributed computing |
| concepts[5].id | https://openalex.org/C2777904410 |
| concepts[5].level | 2 |
| concepts[5].score | 0.46700987219810486 |
| concepts[5].wikidata | https://www.wikidata.org/wiki/Q7397 |
| concepts[5].display_name | Software |
| concepts[6].id | https://openalex.org/C2778774385 |
| concepts[6].level | 3 |
| concepts[6].score | 0.4191368520259857 |
| concepts[6].wikidata | https://www.wikidata.org/wiki/Q4437810 |
| concepts[6].display_name | Power management |
| concepts[7].id | https://openalex.org/C111919701 |
| concepts[7].level | 1 |
| concepts[7].score | 0.2560219168663025 |
| concepts[7].wikidata | https://www.wikidata.org/wiki/Q9135 |
| concepts[7].display_name | Operating system |
| concepts[8].id | https://openalex.org/C163258240 |
| concepts[8].level | 2 |
| concepts[8].score | 0.2391299605369568 |
| concepts[8].wikidata | https://www.wikidata.org/wiki/Q25342 |
| concepts[8].display_name | Power (physics) |
| concepts[9].id | https://openalex.org/C154945302 |
| concepts[9].level | 1 |
| concepts[9].score | 0.08097663521766663 |
| concepts[9].wikidata | https://www.wikidata.org/wiki/Q11660 |
| concepts[9].display_name | Artificial intelligence |
| concepts[10].id | https://openalex.org/C62520636 |
| concepts[10].level | 1 |
| concepts[10].score | 0.0 |
| concepts[10].wikidata | https://www.wikidata.org/wiki/Q944 |
| concepts[10].display_name | Quantum mechanics |
| concepts[11].id | https://openalex.org/C76155785 |
| concepts[11].level | 1 |
| concepts[11].score | 0.0 |
| concepts[11].wikidata | https://www.wikidata.org/wiki/Q418 |
| concepts[11].display_name | Telecommunications |
| concepts[12].id | https://openalex.org/C121332964 |
| concepts[12].level | 0 |
| concepts[12].score | 0.0 |
| concepts[12].wikidata | https://www.wikidata.org/wiki/Q413 |
| concepts[12].display_name | Physics |
| keywords[0].id | https://openalex.org/keywords/computer-science |
| keywords[0].score | 0.8175351023674011 |
| keywords[0].display_name | Computer science |
| keywords[1].id | https://openalex.org/keywords/latency |
| keywords[1].score | 0.7684193849563599 |
| keywords[1].display_name | Latency (audio) |
| keywords[2].id | https://openalex.org/keywords/inference |
| keywords[2].score | 0.5852155089378357 |
| keywords[2].display_name | Inference |
| keywords[3].id | https://openalex.org/keywords/embedded-system |
| keywords[3].score | 0.5152907371520996 |
| keywords[3].display_name | Embedded system |
| keywords[4].id | https://openalex.org/keywords/distributed-computing |
| keywords[4].score | 0.48680922389030457 |
| keywords[4].display_name | Distributed computing |
| keywords[5].id | https://openalex.org/keywords/software |
| keywords[5].score | 0.46700987219810486 |
| keywords[5].display_name | Software |
| keywords[6].id | https://openalex.org/keywords/power-management |
| keywords[6].score | 0.4191368520259857 |
| keywords[6].display_name | Power management |
| keywords[7].id | https://openalex.org/keywords/operating-system |
| keywords[7].score | 0.2560219168663025 |
| keywords[7].display_name | Operating system |
| keywords[8].id | https://openalex.org/keywords/power |
| keywords[8].score | 0.2391299605369568 |
| keywords[8].display_name | Power (physics) |
| keywords[9].id | https://openalex.org/keywords/artificial-intelligence |
| keywords[9].score | 0.08097663521766663 |
| keywords[9].display_name | Artificial intelligence |
| language | en |
| locations[0].id | pmh:oai:eprints.soton.ac.uk:450052 |
| locations[0].is_oa | True |
| locations[0].source.id | https://openalex.org/S4306401019 |
| locations[0].source.issn | |
| locations[0].source.type | repository |
| locations[0].source.is_oa | False |
| locations[0].source.issn_l | |
| locations[0].source.is_core | False |
| locations[0].source.is_in_doaj | False |
| locations[0].source.display_name | ePrints Soton (University of Southampton) |
| locations[0].source.host_organization | https://openalex.org/I43439940 |
| locations[0].source.host_organization_name | University of Southampton |
| locations[0].source.host_organization_lineage | https://openalex.org/I43439940 |
| locations[0].license | |
| locations[0].pdf_url | https://eprints.soton.ac.uk/450052/1/Runtime_DNN_Performance_Scaling_through_Resource_Management_on_Heterogeneous_Embedded_Platforms.pdf |
| locations[0].version | submittedVersion |
| locations[0].raw_type | Conference or Workshop Item |
| locations[0].license_id | |
| locations[0].is_accepted | False |
| locations[0].is_published | False |
| locations[0].raw_source_name | |
| locations[0].landing_page_url | |
| locations[1].id | mag:3181572714 |
| locations[1].is_oa | False |
| locations[1].source | |
| locations[1].license | |
| locations[1].pdf_url | |
| locations[1].version | |
| locations[1].raw_type | |
| locations[1].license_id | |
| locations[1].is_accepted | False |
| locations[1].is_published | |
| locations[1].raw_source_name | |
| locations[1].landing_page_url | https://eprints.soton.ac.uk/450052/ |
| authorships[0].author.id | https://openalex.org/A5091490292 |
| authorships[0].author.orcid | https://orcid.org/0000-0001-5118-9294 |
| authorships[0].author.display_name | Xun Lei |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Lei Xun |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5012783672 |
| authorships[1].author.orcid | https://orcid.org/0000-0002-3591-1328 |
| authorships[1].author.display_name | Bashir M. Al‐Hashimi |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Bashir Al-Hashimi |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5067505586 |
| authorships[2].author.orcid | https://orcid.org/0000-0003-2921-4283 |
| authorships[2].author.display_name | Jonathon Hare |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Jonathon Hare |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A5001556143 |
| authorships[3].author.orcid | https://orcid.org/0000-0003-4980-3894 |
| authorships[3].author.display_name | Geoff V. Merrett |
| authorships[3].author_position | last |
| authorships[3].raw_author_name | Geoff Merrett |
| authorships[3].is_corresponding | False |
| has_content.pdf | True |
| has_content.grobid_xml | False |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://eprints.soton.ac.uk/450052/1/Runtime_DNN_Performance_Scaling_through_Resource_Management_on_Heterogeneous_Embedded_Platforms.pdf |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | Runtime DNN performance scaling through resource management on heterogeneous embedded platforms |
| has_fulltext | True |
| is_retracted | False |
| updated_date | 2025-11-06T04:12:42.849631 |
| primary_topic.id | https://openalex.org/T10502 |
| primary_topic.field.id | https://openalex.org/fields/22 |
| primary_topic.field.display_name | Engineering |
| primary_topic.score | 0.9998999834060669 |
| primary_topic.domain.id | https://openalex.org/domains/3 |
| primary_topic.domain.display_name | Physical Sciences |
| primary_topic.subfield.id | https://openalex.org/subfields/2208 |
| primary_topic.subfield.display_name | Electrical and Electronic Engineering |
| primary_topic.display_name | Advanced Memory and Neural Computing |
| related_works | https://openalex.org/W3123875677, https://openalex.org/W3133149219, https://openalex.org/W3029362297, https://openalex.org/W2554777827, https://openalex.org/W3091175932, https://openalex.org/W2964578286, https://openalex.org/W2945129842, https://openalex.org/W2200587416, https://openalex.org/W2808990982, https://openalex.org/W1970942414, https://openalex.org/W1990339909, https://openalex.org/W2795363269, https://openalex.org/W2381365969, https://openalex.org/W2432370674, https://openalex.org/W2003735, https://openalex.org/W2897519942, https://openalex.org/W2048706569, https://openalex.org/W2983440318, https://openalex.org/W2531302887, https://openalex.org/W2920096131 |
| cited_by_count | 0 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:eprints.soton.ac.uk:450052 |
| best_oa_location.is_oa | True |
| best_oa_location.source.id | https://openalex.org/S4306401019 |
| best_oa_location.source.issn | |
| best_oa_location.source.type | repository |
| best_oa_location.source.is_oa | False |
| best_oa_location.source.issn_l | |
| best_oa_location.source.is_core | False |
| best_oa_location.source.is_in_doaj | False |
| best_oa_location.source.display_name | ePrints Soton (University of Southampton) |
| best_oa_location.source.host_organization | https://openalex.org/I43439940 |
| best_oa_location.source.host_organization_name | University of Southampton |
| best_oa_location.source.host_organization_lineage | https://openalex.org/I43439940 |
| best_oa_location.license | |
| best_oa_location.pdf_url | https://eprints.soton.ac.uk/450052/1/Runtime_DNN_Performance_Scaling_through_Resource_Management_on_Heterogeneous_Embedded_Platforms.pdf |
| best_oa_location.version | submittedVersion |
| best_oa_location.raw_type | Conference or Workshop Item |
| best_oa_location.license_id | |
| best_oa_location.is_accepted | False |
| best_oa_location.is_published | False |
| best_oa_location.raw_source_name | |
| best_oa_location.landing_page_url | |
| primary_location.id | pmh:oai:eprints.soton.ac.uk:450052 |
| primary_location.is_oa | True |
| primary_location.source.id | https://openalex.org/S4306401019 |
| primary_location.source.issn | |
| primary_location.source.type | repository |
| primary_location.source.is_oa | False |
| primary_location.source.issn_l | |
| primary_location.source.is_core | False |
| primary_location.source.is_in_doaj | False |
| primary_location.source.display_name | ePrints Soton (University of Southampton) |
| primary_location.source.host_organization | https://openalex.org/I43439940 |
| primary_location.source.host_organization_name | University of Southampton |
| primary_location.source.host_organization_lineage | https://openalex.org/I43439940 |
| primary_location.license | |
| primary_location.pdf_url | https://eprints.soton.ac.uk/450052/1/Runtime_DNN_Performance_Scaling_through_Resource_Management_on_Heterogeneous_Embedded_Platforms.pdf |
| primary_location.version | submittedVersion |
| primary_location.raw_type | Conference or Workshop Item |
| primary_location.license_id | |
| primary_location.is_accepted | False |
| primary_location.is_published | False |
| primary_location.raw_source_name | |
| primary_location.landing_page_url | |
| publication_date | 2021-06-10 |
| publication_year | 2021 |
| referenced_works | https://openalex.org/W3175549063, https://openalex.org/W3013024145 |
| referenced_works_count | 2 |
| abstract_inverted_index.a | 24 |
| abstract_inverted_index.In | 51 |
| abstract_inverted_index.at | 78 |
| abstract_inverted_index.be | 60 |
| abstract_inverted_index.by | 69 |
| abstract_inverted_index.in | 15 |
| abstract_inverted_index.is | 2, 33 |
| abstract_inverted_index.it | 32 |
| abstract_inverted_index.of | 26, 66 |
| abstract_inverted_index.on | 7, 90 |
| abstract_inverted_index.to | 11, 35, 46, 62 |
| abstract_inverted_index.we | 54 |
| abstract_inverted_index.DNN | 0, 48, 67 |
| abstract_inverted_index.and | 18, 28, 71, 75 |
| abstract_inverted_index.can | 59 |
| abstract_inverted_index.due | 10 |
| abstract_inverted_index.how | 56, 83 |
| abstract_inverted_index.the | 12, 41, 47, 64 |
| abstract_inverted_index.DNNs | 85 |
| abstract_inverted_index.SoCs | 21 |
| abstract_inverted_index.This | 80 |
| abstract_inverted_index.both | 73 |
| abstract_inverted_index.meet | 37 |
| abstract_inverted_index.show | 55 |
| abstract_inverted_index.this | 52 |
| abstract_inverted_index.vary | 49 |
| abstract_inverted_index.with | 88 |
| abstract_inverted_index.work | 81 |
| abstract_inverted_index.being | 4 |
| abstract_inverted_index.clear | 13 |
| abstract_inverted_index.local | 42 |
| abstract_inverted_index.shows | 82 |
| abstract_inverted_index.Modern | 20 |
| abstract_inverted_index.tuning | 72 |
| abstract_inverted_index.CPU-GPU | 93 |
| abstract_inverted_index.applied | 61 |
| abstract_inverted_index.because | 40 |
| abstract_inverted_index.budgets | 39 |
| abstract_inverted_index.dynamic | 29, 84 |
| abstract_inverted_index.execute | 23 |
| abstract_inverted_index.locally | 6 |
| abstract_inverted_index.poster, | 53 |
| abstract_inverted_index.privacy | 17 |
| abstract_inverted_index.accuracy | 87 |
| abstract_inverted_index.embedded | 8, 92 |
| abstract_inverted_index.executed | 5 |
| abstract_inverted_index.hardware | 76 |
| abstract_inverted_index.latency, | 16 |
| abstract_inverted_index.optimise | 63 |
| abstract_inverted_index.resource | 57 |
| abstract_inverted_index.runtime. | 79 |
| abstract_inverted_index.software | 74 |
| abstract_inverted_index.available | 45 |
| abstract_inverted_index.computing | 43 |
| abstract_inverted_index.different | 27 |
| abstract_inverted_index.inference | 1 |
| abstract_inverted_index.platform. | 94 |
| abstract_inverted_index.resources | 44 |
| abstract_inverted_index.trade-off | 86 |
| abstract_inverted_index.typically | 22 |
| abstract_inverted_index.workloads | 30, 68 |
| abstract_inverted_index.advantages | 14 |
| abstract_inverted_index.constantly | 77 |
| abstract_inverted_index.management | 58 |
| abstract_inverted_index.monitoring | 70 |
| abstract_inverted_index.platforms, | 9 |
| abstract_inverted_index.challenging | 34 |
| abstract_inverted_index.combination | 25 |
| abstract_inverted_index.performance | 65 |
| abstract_inverted_index.consistently | 36 |
| abstract_inverted_index.increasingly | 3 |
| abstract_inverted_index.concurrently, | 31 |
| abstract_inverted_index.connectivity. | 19 |
| abstract_inverted_index.considerably. | 50 |
| abstract_inverted_index.heterogeneous | 91 |
| abstract_inverted_index.latency/energy | 38 |
| abstract_inverted_index.latency/energy/power | 89 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 4 |
| sustainable_development_goals[0].id | https://metadata.un.org/sdg/8 |
| sustainable_development_goals[0].score | 0.44999998807907104 |
| sustainable_development_goals[0].display_name | Decent work and economic growth |
| citation_normalized_percentile.value | 0.0662896 |
| citation_normalized_percentile.is_in_top_1_percent | False |
| citation_normalized_percentile.is_in_top_10_percent | False |