Opportunities of Renewable Energy Powered DNN Inference
With the proliferation of the adoption of renewable energy in powering data centers, addressing the challenges of such energy sources has attracted researchers from academia and industry. One of the challenging characteristics of data centers with renewable energy is the intrinsic power fluctuation....
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | |
|---|---|
| container_issue | |
| container_start_page | |
| container_title | |
| container_volume | |
| creator | Nabavinejad, Seyed Morteza Guo, Tian |
| description | With the proliferation of the adoption of renewable energy in powering data
centers, addressing the challenges of such energy sources has attracted
researchers from academia and industry. One of the challenging characteristics
of data centers with renewable energy is the intrinsic power fluctuation.
Fluctuation in renewable power supply inevitably requires adjusting
applications' power consumption, which can lead to undesirable performance
degradation. This paper investigates the possible control knobs to manage the
power and performance of a popular cloud workload, i.e., deep neural network
inference, under the fluctuating power supply. Through empirical profiling and
trace-driven simulations, we observe the different impact levels associated
with inference control knobs on throughput, under varying power supplies. Based
on our observations, we provide a list of future research directions to
leverage the control knobs to achieve high throughput. |
| doi_str_mv | 10.48550/arxiv.2306.12247 |
| format | Article |
| fullrecord | <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2306_12247</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2306_12247</sourcerecordid><originalsourceid>FETCH-LOGICAL-a677-7ca5f282d71f4c378cd430773317353e6dda99c0b9e09f6fb8b5762c732e2f6c3</originalsourceid><addsrcrecordid>eNotj8luwjAURb1hgYAPYFX_QFLHL_ZLlohZQlBV7CPHfq4iUScylOHvmbo6d3V0D2PjTKR5oZT4NPHanFMJQqeZlDn2Ge66ro2nv9CcGjry1vNvCnQx9YH4PFD8ufGv9kKRHJ9tt3wd_GMHS0PW8-ZwpNE_B2y_mO-nq2SzW66nk01iNGKC1igvC-kw87kFLKzLQSACZAgKSDtnytKKuiRReu3rolaopUWQJL22MGAfb-3redXF5tfEW_UsqF4FcAcWfD_-</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Opportunities of Renewable Energy Powered DNN Inference</title><source>arXiv.org</source><creator>Nabavinejad, Seyed Morteza ; Guo, Tian</creator><creatorcontrib>Nabavinejad, Seyed Morteza ; Guo, Tian</creatorcontrib><description>With the proliferation of the adoption of renewable energy in powering data
centers, addressing the challenges of such energy sources has attracted
researchers from academia and industry. One of the challenging characteristics
of data centers with renewable energy is the intrinsic power fluctuation.
Fluctuation in renewable power supply inevitably requires adjusting
applications' power consumption, which can lead to undesirable performance
degradation. This paper investigates the possible control knobs to manage the
power and performance of a popular cloud workload, i.e., deep neural network
inference, under the fluctuating power supply. Through empirical profiling and
trace-driven simulations, we observe the different impact levels associated
with inference control knobs on throughput, under varying power supplies. Based
on our observations, we provide a list of future research directions to
leverage the control knobs to achieve high throughput.</description><identifier>DOI: 10.48550/arxiv.2306.12247</identifier><language>eng</language><subject>Computer Science - Distributed, Parallel, and Cluster Computing</subject><creationdate>2023-06</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2306.12247$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2306.12247$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Nabavinejad, Seyed Morteza</creatorcontrib><creatorcontrib>Guo, Tian</creatorcontrib><title>Opportunities of Renewable Energy Powered DNN Inference</title><description>With the proliferation of the adoption of renewable energy in powering data
centers, addressing the challenges of such energy sources has attracted
researchers from academia and industry. One of the challenging characteristics
of data centers with renewable energy is the intrinsic power fluctuation.
Fluctuation in renewable power supply inevitably requires adjusting
applications' power consumption, which can lead to undesirable performance
degradation. This paper investigates the possible control knobs to manage the
power and performance of a popular cloud workload, i.e., deep neural network
inference, under the fluctuating power supply. Through empirical profiling and
trace-driven simulations, we observe the different impact levels associated
with inference control knobs on throughput, under varying power supplies. Based
on our observations, we provide a list of future research directions to
leverage the control knobs to achieve high throughput.</description><subject>Computer Science - Distributed, Parallel, and Cluster Computing</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotj8luwjAURb1hgYAPYFX_QFLHL_ZLlohZQlBV7CPHfq4iUScylOHvmbo6d3V0D2PjTKR5oZT4NPHanFMJQqeZlDn2Ge66ro2nv9CcGjry1vNvCnQx9YH4PFD8ufGv9kKRHJ9tt3wd_GMHS0PW8-ZwpNE_B2y_mO-nq2SzW66nk01iNGKC1igvC-kw87kFLKzLQSACZAgKSDtnytKKuiRReu3rolaopUWQJL22MGAfb-3redXF5tfEW_UsqF4FcAcWfD_-</recordid><startdate>20230621</startdate><enddate>20230621</enddate><creator>Nabavinejad, Seyed Morteza</creator><creator>Guo, Tian</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20230621</creationdate><title>Opportunities of Renewable Energy Powered DNN Inference</title><author>Nabavinejad, Seyed Morteza ; Guo, Tian</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a677-7ca5f282d71f4c378cd430773317353e6dda99c0b9e09f6fb8b5762c732e2f6c3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Computer Science - Distributed, Parallel, and Cluster Computing</topic><toplevel>online_resources</toplevel><creatorcontrib>Nabavinejad, Seyed Morteza</creatorcontrib><creatorcontrib>Guo, Tian</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Nabavinejad, Seyed Morteza</au><au>Guo, Tian</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Opportunities of Renewable Energy Powered DNN Inference</atitle><date>2023-06-21</date><risdate>2023</risdate><abstract>With the proliferation of the adoption of renewable energy in powering data
centers, addressing the challenges of such energy sources has attracted
researchers from academia and industry. One of the challenging characteristics
of data centers with renewable energy is the intrinsic power fluctuation.
Fluctuation in renewable power supply inevitably requires adjusting
applications' power consumption, which can lead to undesirable performance
degradation. This paper investigates the possible control knobs to manage the
power and performance of a popular cloud workload, i.e., deep neural network
inference, under the fluctuating power supply. Through empirical profiling and
trace-driven simulations, we observe the different impact levels associated
with inference control knobs on throughput, under varying power supplies. Based
on our observations, we provide a list of future research directions to
leverage the control knobs to achieve high throughput.</abstract><doi>10.48550/arxiv.2306.12247</doi><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | DOI: 10.48550/arxiv.2306.12247 |
| ispartof | |
| issn | |
| language | eng |
| recordid | cdi_arxiv_primary_2306_12247 |
| source | arXiv.org |
| subjects | Computer Science - Distributed, Parallel, and Cluster Computing |
| title | Opportunities of Renewable Energy Powered DNN Inference |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-27T05%3A19%3A11IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Opportunities%20of%20Renewable%20Energy%20Powered%20DNN%20Inference&rft.au=Nabavinejad,%20Seyed%20Morteza&rft.date=2023-06-21&rft_id=info:doi/10.48550/arxiv.2306.12247&rft_dat=%3Carxiv_GOX%3E2306_12247%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |