CILP: An Arbitrary-bit Precision All-digital Compute-in-memory Solver for Integer Linear Programming Problems

Integer Linear Programming (ILP) is an NP-complete combinatorial optimization problem (COP), suggesting that it is computationally challenging to solve due to its exponentially increased operations with scaling. As shown in Fig. 1, The ILP is relevant in various real-world scenarios such as computat...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings of the Custom Integrated Circuits Conference S. 1 - 2
Hauptverfasser: Yang, Mengtian, Wang, Yipeng, Xie, Shanshan, Lo, Chieh-Pu, Wang, Meizhi, Oruganti, Sirish, Sehgal, Rishabh, Kulkarni, Jaydeep P.
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 21.04.2024
Schlagworte:
ISSN:2152-3630
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Integer Linear Programming (ILP) is an NP-complete combinatorial optimization problem (COP), suggesting that it is computationally challenging to solve due to its exponentially increased operations with scaling. As shown in Fig. 1, The ILP is relevant in various real-world scenarios such as computational biology [1], investment decision, automated driving, and electronic design automation [2]. An ILP solver aims to find a set of integer variables (x) to maximize a linear objective function (c\cdot x) , subject to a set of linear constraints (A\cdot x\leq b) . With the increasingly wide usage of ILP, various new solving algorithms [3] have been proposed, but the performance are limited by substantial memory access. ILP coefficients are fixed during solving, but software solvers on cache-register architectures frequently access cache to reload coefficients because of small register file size, causing up to a 10^{14}\mathrm{x} disparity between stored and accessed memory bits. FPGA [4] and AISC [5] accelerated solvers improve the speed by customized processing element (PE), but they still need frequent accesses to Block-RAM or scratch pad. Compute-in-memory (CIM) solutions are well-suited for ILP solving which has extremely high data reuse, but existing CIM DNN accelerators incur precision loss with hardware tradeoffs, which is unacceptable for ILP where the feasibility checking must be correct. Previous all-digital CIM COP solver for Boolean variables [6] uses a customized 6\mathrm{T} -6T{###} 3\mathrm{T} cell, limiting their adaptability to different technologies.
AbstractList Integer Linear Programming (ILP) is an NP-complete combinatorial optimization problem (COP), suggesting that it is computationally challenging to solve due to its exponentially increased operations with scaling. As shown in Fig. 1, The ILP is relevant in various real-world scenarios such as computational biology [1], investment decision, automated driving, and electronic design automation [2]. An ILP solver aims to find a set of integer variables (x) to maximize a linear objective function (c\cdot x) , subject to a set of linear constraints (A\cdot x\leq b) . With the increasingly wide usage of ILP, various new solving algorithms [3] have been proposed, but the performance are limited by substantial memory access. ILP coefficients are fixed during solving, but software solvers on cache-register architectures frequently access cache to reload coefficients because of small register file size, causing up to a 10^{14}\mathrm{x} disparity between stored and accessed memory bits. FPGA [4] and AISC [5] accelerated solvers improve the speed by customized processing element (PE), but they still need frequent accesses to Block-RAM or scratch pad. Compute-in-memory (CIM) solutions are well-suited for ILP solving which has extremely high data reuse, but existing CIM DNN accelerators incur precision loss with hardware tradeoffs, which is unacceptable for ILP where the feasibility checking must be correct. Previous all-digital CIM COP solver for Boolean variables [6] uses a customized 6\mathrm{T} -6T{###} 3\mathrm{T} cell, limiting their adaptability to different technologies.
Author Kulkarni, Jaydeep P.
Wang, Yipeng
Lo, Chieh-Pu
Yang, Mengtian
Oruganti, Sirish
Xie, Shanshan
Wang, Meizhi
Sehgal, Rishabh
Author_xml – sequence: 1
  givenname: Mengtian
  surname: Yang
  fullname: Yang, Mengtian
  organization: The University of Texas at Austin,Austin,TX
– sequence: 2
  givenname: Yipeng
  surname: Wang
  fullname: Wang, Yipeng
  organization: The University of Texas at Austin,Austin,TX
– sequence: 3
  givenname: Shanshan
  surname: Xie
  fullname: Xie, Shanshan
  organization: The University of Texas at Austin,Austin,TX
– sequence: 4
  givenname: Chieh-Pu
  surname: Lo
  fullname: Lo, Chieh-Pu
  organization: The University of Texas at Austin,Austin,TX
– sequence: 5
  givenname: Meizhi
  surname: Wang
  fullname: Wang, Meizhi
  organization: The University of Texas at Austin,Austin,TX
– sequence: 6
  givenname: Sirish
  surname: Oruganti
  fullname: Oruganti, Sirish
  organization: The University of Texas at Austin,Austin,TX
– sequence: 7
  givenname: Rishabh
  surname: Sehgal
  fullname: Sehgal, Rishabh
  organization: The University of Texas at Austin,Austin,TX
– sequence: 8
  givenname: Jaydeep P.
  surname: Kulkarni
  fullname: Kulkarni, Jaydeep P.
  organization: The University of Texas at Austin,Austin,TX
BookMark eNo1kN9KwzAchaMouM29gWBfIPOXpMkS70bxT6HgQL0eaZOUSJOMtAp7eyvq1fk4Fx-cs0QXMUWL0C2BDSGg7qq6qgQorjYUaLkhwKmCkpyhtdoqyTgwVYIg52hBCaeYCQZXaDmOHwBEKUkXKFR1s78vdrHY5dZPWecTnrPYZ9v50ae5HwZsfO8nPRRVCsfPyWIfcbAh5VPxmoYvmwuXclHHyfYzNz5anWdD6rMOwcf-h9vBhvEaXTo9jHb9lyv0_vjwVj3j5uWprnYN9hTKCZtOdbIVWkC5dVQaUkILmkmrteElY86BELzTxoC0kmuqTMtAOSEEM-AkW6GbX6-31h6O2Yd51-H_HvYN4FFctw
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1109/CICC60959.2024.10529041
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan (POP) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Xplore
IEEE Proceedings Order Plans (POP) 1998-present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Xplore
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
EISBN 9798350394061
EISSN 2152-3630
EndPage 2
ExternalDocumentID 10529041
Genre orig-research
GrantInformation_xml – fundername: TSMC
  funderid: 10.13039/501100004368
GroupedDBID 29O
6IE
6IF
6IH
6IK
6IL
6IM
6IN
AAJGR
AAWTH
ABLEC
ACGFS
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IEGSK
IJVOP
IPLJI
M43
OCL
RIE
RIL
RIO
RNS
ID FETCH-LOGICAL-i204t-dc9c8b6a6047f28d140b0a38eaad5433ff0665cadd08e85a29db309f6663d0f83
IEDL.DBID RIE
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001230023800081&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Wed Aug 27 02:05:34 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i204t-dc9c8b6a6047f28d140b0a38eaad5433ff0665cadd08e85a29db309f6663d0f83
PageCount 2
ParticipantIDs ieee_primary_10529041
PublicationCentury 2000
PublicationDate 2024-April-21
PublicationDateYYYYMMDD 2024-04-21
PublicationDate_xml – month: 04
  year: 2024
  text: 2024-April-21
  day: 21
PublicationDecade 2020
PublicationTitle Proceedings of the Custom Integrated Circuits Conference
PublicationTitleAbbrev CICC
PublicationYear 2024
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0019982
Score 2.2679052
Snippet Integer Linear Programming (ILP) is an NP-complete combinatorial optimization problem (COP), suggesting that it is computationally challenging to solve due to...
SourceID ieee
SourceType Publisher
StartPage 1
SubjectTerms Common Information Model (computing)
Computer architecture
In-memory computing
Integer linear programming
Linear programming
Microprocessors
Software algorithms
Title CILP: An Arbitrary-bit Precision All-digital Compute-in-memory Solver for Integer Linear Programming Problems
URI https://ieeexplore.ieee.org/document/10529041
WOSCitedRecordID wos001230023800081&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07a8MwEBZt6NAufaX0jYauSmVJtqVuwTQ0EEKgD7IFyZKDIbFL4hT676tznLQdOnQ7DOJAhyR_d_fdh9AdXIAgFEecNJQIJiQx2gPXCIIvrDaszne8DeLhUI7HatSQ1WsujHOubj5zHTDrWr4t0xWkyvwJD5miQFPfjeN4Tdbalgw8bmBNA1dA1X3STxIYpgZkFCY6m6W_RFTqN6R3-E_vR6j9zcbDo-07c4x2XHGCDn4MEjxF86Q_GD3gboG7C5PXTHoPeSu_rJHQwd3ZjNh8ChIhuFFyIHlB5tBo-4mfS2iQxv4HFkOKcOptD1L9IQDH0L81927ABvGZZRu99h5fkifSCCmQnFFREZuqVJpIR1TEGZPWgypDNZdOaxsKzrMMCjCpv-qodDLUTFnDqco8tOGWZpKfoVZRFu4c4UDJCASPLKNGRNr4MIdxqngWW8cDEV6gNuzc5H09K2Oy2bTLP75foX2ID9RnWHCNWtVi5W7QXvpR5cvFbR3hLxB_ptU
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3PS8MwFA4yBfXir4m_zcFrZpqkbeJtFMeGcwycsttomnQUtk62TvC_N6_rph48eHsUwoM8kvR7733vQ-gOLkAQiiNWakoEE5Lo2AHXAIIvTKxZme9464a9nhwOVb8iq5dcGGtt2XxmG2CWtXwzS5aQKnMn3GeKAk192xeCeSu61qZo4JADq1q4PKruo04UwTg1oKMw0Vgv_iWjUr4irYN_-j9E9W8-Hu5vXpojtGXzY7T_Y5TgCZpGnW7_ATdz3JzrrOTSO9BbuGWViA5uTibEZGMQCcGVlgPJcjKFVttP_DKDFmnsfmExJAnHznYw1R0DcAwdXFPnBmyQn1nU0WvrcRC1SSWlQDJGRUFMohKpgzigIkyZNA5WaRpzaePY-ILzNIUSTOIuOyqt9GOmjOZUpQ7ccENTyU9RLZ_l9gxhT8kAJI8Mo1oEsXaB9sNE8TQ0lnvCP0d12LnR-2paxmi9aRd_fL9Fu-3Bc3fU7fSeLtEexAqqNcy7QrVivrTXaCf5KLLF_KaM9hfqq6oc
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%3Abook&rft.genre=proceeding&rft.title=Proceedings+of+the+Custom+Integrated+Circuits+Conference&rft.atitle=CILP%3A+An+Arbitrary-bit+Precision+All-digital+Compute-in-memory+Solver+for+Integer+Linear+Programming+Problems&rft.au=Yang%2C+Mengtian&rft.au=Wang%2C+Yipeng&rft.au=Xie%2C+Shanshan&rft.au=Lo%2C+Chieh-Pu&rft.date=2024-04-21&rft.pub=IEEE&rft.eissn=2152-3630&rft.spage=1&rft.epage=2&rft_id=info:doi/10.1109%2FCICC60959.2024.10529041&rft.externalDocID=10529041