Power-mode-aware Memory Subsystem Optimization for Low-power System-on-Chip Design
The memory subsystem is increasingly subject to an intensive energy minimization effort in embedded and System-on-Chip development. While the main focus is typically put on energy consumption reduction, there are other optimization aspects that become more and more relevant as well, e.g., peak power...
Saved in:
Published in | ACM transactions on embedded computing systems Vol. 18; no. 5; pp. 1 - 25 |
---|---|
Main Authors | , |
Format | Journal Article |
Language | English |
Published |
01.10.2019
|
Online Access | Get full text |
ISSN | 1539-9087 1558-3465 |
DOI | 10.1145/3356583 |
Cover
Loading…
Abstract | The memory subsystem is increasingly subject to an intensive energy minimization effort in embedded and System-on-Chip development. While the main focus is typically put on energy consumption reduction, there are other optimization aspects that become more and more relevant as well, e.g., peak power constraints or time budgets. In this regard, the present article makes the following contributions. Taking industrial-grade information into account, different Static Random-Access Memory (SRAM) power modes and their characteristics are presented at first. Using this information, a comprehensive optimization model with the main intention of energy minimization is defined. It is based on memory access statistics that represent the embedded software of interest, which allows for application-tailored improvements. Further, it considers different power states of the memory subsystem and enables the definition of peak power and time corridor constraints. The presented two-stage implementation of this optimization model allows the handling of large design spaces. Clearly defined interfaces facilitate the exchange of individual workflow parts in a plug-and-play fashion and further enable a neat integration of our optimization method with existing hardware/software (HW/SW) codesign synthesis flows. A general evaluation for different technology nodes yields that the optimization potential of memory low-power modes increases with advancing miniaturization but also depends on the data footprint of the embedded software. Experimental results for a set of benchmark applications confirm these findings and provide energy savings of up to 90% and over 60% on average compared to a monolithic memory layout without low-power modes. |
---|---|
AbstractList | The memory subsystem is increasingly subject to an intensive energy minimization effort in embedded and System-on-Chip development. While the main focus is typically put on energy consumption reduction, there are other optimization aspects that become more and more relevant as well, e.g., peak power constraints or time budgets. In this regard, the present article makes the following contributions. Taking industrial-grade information into account, different Static Random-Access Memory (SRAM) power modes and their characteristics are presented at first. Using this information, a comprehensive optimization model with the main intention of energy minimization is defined. It is based on memory access statistics that represent the embedded software of interest, which allows for application-tailored improvements. Further, it considers different power states of the memory subsystem and enables the definition of peak power and time corridor constraints. The presented two-stage implementation of this optimization model allows the handling of large design spaces. Clearly defined interfaces facilitate the exchange of individual workflow parts in a plug-and-play fashion and further enable a neat integration of our optimization method with existing hardware/software (HW/SW) codesign synthesis flows. A general evaluation for different technology nodes yields that the optimization potential of memory low-power modes increases with advancing miniaturization but also depends on the data footprint of the embedded software. Experimental results for a set of benchmark applications confirm these findings and provide energy savings of up to 90% and over 60% on average compared to a monolithic memory layout without low-power modes. |
Author | Radetzki, Martin Strobel, Manuel |
Author_xml | – sequence: 1 givenname: Manuel orcidid: 0000-0003-4770-4147 surname: Strobel fullname: Strobel, Manuel organization: University of Stuttgart, Pfaffenwaldring, Stuttgart, Germany – sequence: 2 givenname: Martin surname: Radetzki fullname: Radetzki, Martin organization: University of Stuttgart, Pfaffenwaldring, Stuttgart, Germany |
BookMark | eNplkMtOwzAURC1UJNqC-AXvWBn8iB9ZovKUiooorCPHuQGjJo7soKp8PUnpClYzi3OvZmaGJm1oAaFzRi8Zy-SVEFJJI47QlElpiMiUnIxe5CSnRp-gWUqflDLNMzlFL89hC5E0oQJitzYCfoImxB1ef5Vpl3po8KrrfeO_be9Di-sQ8TJsSTee4fWeIKEliw_f4RtI_r09Rce13SQ4O-gcvd3dvi4eyHJ1_7i4XhLHVdYPYaThGWeQg1Y0p5zyymlnFVQaJKVD8FowYRQ1Qy2nRVU5pa3krCprZkoxRxe_f10MKUWoiy76xsZdwWgxTlEcphhI8od0vt_36aP1m3_8DwisYJI |
CitedBy_id | crossref_primary_10_1007_s13204_021_02098_7 |
Cites_doi | 10.1109/TC.2010.43 10.1109/TVLSI.2008.2001940 10.1145/2560019 10.1109/DATE.2002.998306 10.1145/1124713.1124725 10.1109/SOCC.2005.1554478 10.1109/TVLSI.2005.859478 10.1109/TCAD.2005.852299 10.1109/WISES.2017.7986927 10.1145/581888.581891 10.1080/15427951.2004.10129093 10.1007/978-3-642-38853-8_7 10.1109/ICSAMOS.2006.300807 10.1103/PhysRevE.69.026113 10.1145/344166.344518 10.1109/TCAD.2005.859521 10.1049/mnl.2011.0680 |
ContentType | Journal Article |
DBID | AAYXX CITATION |
DOI | 10.1145/3356583 |
DatabaseName | CrossRef |
DatabaseTitle | CrossRef |
DatabaseTitleList | CrossRef |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Engineering |
EISSN | 1558-3465 |
EndPage | 25 |
ExternalDocumentID | 10_1145_3356583 |
GroupedDBID | -DZ -~X .4S .DC 23M 4.4 5GY 5VS 6J9 8US AAKMM AALFJ AAYFX AAYXX ABPPZ ACGFO ACM ADBCU ADL ADMLS AEBYY AEFXT AEGXH AEJOY AENEX AENSD AFWIH AFWXC AIAGR AIKLT AKRVB ALMA_UNASSIGNED_HOLDINGS ARCSS ASPBG AVWKF BDXCO CCLIF CITATION CS3 D0L EBS EDO FEDTE GUFHI HGAVV H~9 I07 LHSKQ P1C P2P PQQKQ RNS ROL TUS UPT ZCA |
ID | FETCH-LOGICAL-c264t-90582421e9e76090202dc7ca6ed7e500346f3138608145c73ddc67a521dbf18b3 |
ISSN | 1539-9087 |
IngestDate | Thu Jul 03 08:32:37 EDT 2025 Thu Apr 24 22:57:14 EDT 2025 |
IsPeerReviewed | true |
IsScholarly | true |
Issue | 5 |
Language | English |
LinkModel | OpenURL |
MergedId | FETCHMERGED-LOGICAL-c264t-90582421e9e76090202dc7ca6ed7e500346f3138608145c73ddc67a521dbf18b3 |
ORCID | 0000-0003-4770-4147 |
PageCount | 25 |
ParticipantIDs | crossref_primary_10_1145_3356583 crossref_citationtrail_10_1145_3356583 |
PublicationCentury | 2000 |
PublicationDate | 2019-10-01 |
PublicationDateYYYYMMDD | 2019-10-01 |
PublicationDate_xml | – month: 10 year: 2019 text: 2019-10-01 day: 01 |
PublicationDecade | 2010 |
PublicationTitle | ACM transactions on embedded computing systems |
PublicationYear | 2019 |
References | NanGate Inc. 2011. (e_1_2_1_21_1) 2019 Hu T. C. (e_1_2_1_8_1) 2016 e_1_2_1_23_1 Strobel M. (e_1_2_1_26_1) 2016; 2017 e_1_2_1_24_1 e_1_2_1_22_1 Lin W.-C. (e_1_2_1_12_1) 2004; 2 e_1_2_1_27_1 e_1_2_1_28_1 e_1_2_1_25_1 More Moore ITRS. (e_1_2_1_9_1) 2019 Luz V. De La (e_1_2_1_14_1) 2002 Minwell L. (e_1_2_1_18_1) 2019 e_1_2_1_7_1 e_1_2_1_3_1 Muralimanohar N. (e_1_2_1_20_1) 2009 e_1_2_1_4_1 e_1_2_1_13_1 e_1_2_1_1_1 e_1_2_1_10_1 e_1_2_1_2_1 e_1_2_1_11_1 e_1_2_1_16_1 e_1_2_1_17_1 e_1_2_1_15_1 Guthaus M. R. (e_1_2_1_6_1) 2001 e_1_2_1_19_1 |
References_xml | – ident: e_1_2_1_13_1 doi: 10.1109/TC.2010.43 – ident: e_1_2_1_17_1 doi: 10.1109/TVLSI.2008.2001940 – ident: e_1_2_1_7_1 doi: 10.1145/2560019 – ident: e_1_2_1_25_1 doi: 10.1109/DATE.2002.998306 – ident: e_1_2_1_19_1 doi: 10.1145/1124713.1124725 – volume-title: Retrieved year: 2019 ident: e_1_2_1_9_1 – ident: e_1_2_1_16_1 – ident: e_1_2_1_23_1 doi: 10.1109/SOCC.2005.1554478 – ident: e_1_2_1_10_1 doi: 10.1109/TVLSI.2005.859478 – ident: e_1_2_1_1_1 doi: 10.1109/TCAD.2005.852299 – volume-title: Kahng year: 2016 ident: e_1_2_1_8_1 – ident: e_1_2_1_27_1 doi: 10.1109/WISES.2017.7986927 – ident: e_1_2_1_2_1 doi: 10.1145/581888.581891 – volume: 2017 year: 2016 ident: e_1_2_1_26_1 article-title: Low power memory allocation and mapping for area-constrained systems-on-chips publication-title: EURASIP Journal on Embedded Systems – ident: e_1_2_1_4_1 doi: 10.1080/15427951.2004.10129093 – volume-title: Technical Report HPL-2009-85. HP Laboratories. Retrieved year: 2009 ident: e_1_2_1_20_1 – ident: e_1_2_1_24_1 doi: 10.1007/978-3-642-38853-8_7 – ident: e_1_2_1_11_1 doi: 10.1109/ICSAMOS.2006.300807 – volume: 2 volume-title: Proceedings of the 2004 IEEE Asia-Pacific Conference on Circuits and Systems year: 2004 ident: e_1_2_1_12_1 – volume-title: Proceedings of the 2002 Design Automation Conference. 213--218 year: 2002 ident: e_1_2_1_14_1 – ident: e_1_2_1_22_1 doi: 10.1103/PhysRevE.69.026113 – volume-title: NanGate FreePDK45 Open Cell Library. Retrieved year: 2019 ident: e_1_2_1_21_1 – ident: e_1_2_1_3_1 doi: 10.1145/344166.344518 – ident: e_1_2_1_15_1 doi: 10.1109/TCAD.2005.859521 – volume-title: Proceedings of the 2001 IEEE International Workshop on Workload Characterization. IEEE. DOI:https://doi.org/10 year: 2001 ident: e_1_2_1_6_1 – volume-title: Advanced Power Management in Embedded Memory Subsystems. Retrieved year: 2019 ident: e_1_2_1_18_1 – ident: e_1_2_1_28_1 doi: 10.1049/mnl.2011.0680 |
SSID | ssj0017245 |
Score | 2.1690364 |
Snippet | The memory subsystem is increasingly subject to an intensive energy minimization effort in embedded and System-on-Chip development. While the main focus is... |
SourceID | crossref |
SourceType | Enrichment Source Index Database |
StartPage | 1 |
Title | Power-mode-aware Memory Subsystem Optimization for Low-power System-on-Chip Design |
Volume | 18 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV3JTsMwELWgvcABsYpdPiAuyNA0sZ0cqwKqEAVUQOqtchznBGkFqZD4esZLnVCQWC5R5dpVm5mMx5037yF0pBi8k2eU0DyAAwoTCTxSVBIplEjivJWLVFd0-zes9xhdDemwoicw3SVleirfv-0r-Y9VYQzsqrtk_2BZ_6EwAK_BvnAFC8P1Vza-0xJnRIvZEPGmIVx9jZu10cAwNJ_cQkR4dq2WBlF4PX4jE73McZWTcUG6GrB1XkE5ZqS03b5WkJjJiZu6gnpOFYQq0wo3mRrI9GuN89wAhHWD0ZPtAyqmyiM4BiJT5btVybbkBfV_HILEY9eqIAkWbbmNUrkxGpMwsroPXyNrVbs2YTKo7be27_lrJI806UUYQsJplW4-c2XP7WEeWWj7rOnILVxEzTacHyBiNzvn_et7X2DibaNf7X-K7afWS8_c0lqiUss4HlbRijsq4I61-xpaUMU6Wq4RSG6gwbwHYOsB2HsArnsABg_A3gPwZw_A1gM20ePlxUO3R5xKBpGQzJbw7Wms6_oqUZxplG2rnUkuBVMZV1TzD7E8DMKYQfIXUcnDLJOMC0jbsjQP4jTcQo1iXKhthAVkzzHnSU5lHqUsFJEQ8PRKATumUoLvoOPZLRlJRyGvlUyeRnO3fQdhP3FiWVPmp-z-PGUPLVXut48a5ctUHUAKWKaHzpwfMENfXg |
linkProvider | EBSCOhost |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Power-mode-aware+Memory+Subsystem+Optimization+for+Low-power+System-on-Chip+Design&rft.jtitle=ACM+transactions+on+embedded+computing+systems&rft.au=Strobel%2C+Manuel&rft.au=Radetzki%2C+Martin&rft.date=2019-10-01&rft.issn=1539-9087&rft.eissn=1558-3465&rft.volume=18&rft.issue=5&rft.spage=1&rft.epage=25&rft_id=info:doi/10.1145%2F3356583&rft.externalDBID=n%2Fa&rft.externalDocID=10_1145_3356583 |
thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1539-9087&client=summon |
thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1539-9087&client=summon |
thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1539-9087&client=summon |