PreSto: An In-Storage Data Preprocessing System for Training Recommendation Models

Training recommendation systems (RecSys) faces several challenges as it requires the "data preprocessing" stage to preprocess an ample amount of raw data and feed them to the GPU for training in a seamless manner. To sustain high training throughput, state-of-the-art solutions reserve a la...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:2024 ACM/IEEE 51st Annual International Symposium on Computer Architecture (ISCA) s. 340 - 353
Hlavní autori: Lee, Yunjae, Kim, Hyeseong, Rhu, Minsoo
Médium: Konferenčný príspevok..
Jazyk:English
Vydavateľské údaje: IEEE 29.06.2024
Predmet:
computational storage device
Costs
Data preprocessing
Graphics processing units
near data processing
neural network
Parallel processing
Power demand
Recommendation system
Throughput
Training
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract Training recommendation systems (RecSys) faces several challenges as it requires the "data preprocessing" stage to preprocess an ample amount of raw data and feed them to the GPU for training in a seamless manner. To sustain high training throughput, state-of-the-art solutions reserve a large fleet of CPU servers for preprocessing which incurs substantial deployment cost and power consumption. Our characterization reveals that prior CPU-centric preprocessing is bottlenecked on feature generation and feature normalization operations as it fails to reap out the abundant inter-/intra-feature parallelism in RecSys preprocessing. PreSto is a storage-centric preprocessing system leveraging In-Storage Processing (ISP), which offloads the bottlenecked preprocessing operations to our ISP units. We show that PreSto outperforms the baseline CPU-centric system with a 9.6× speedup in end-to-end preprocessing time, 4.3× enhancement in cost-efficiency, and 11.3× improvement in energy-efficiency on average for production-scale RecSys preprocessing.
AbstractList Training recommendation systems (RecSys) faces several challenges as it requires the "data preprocessing" stage to preprocess an ample amount of raw data and feed them to the GPU for training in a seamless manner. To sustain high training throughput, state-of-the-art solutions reserve a large fleet of CPU servers for preprocessing which incurs substantial deployment cost and power consumption. Our characterization reveals that prior CPU-centric preprocessing is bottlenecked on feature generation and feature normalization operations as it fails to reap out the abundant inter-/intra-feature parallelism in RecSys preprocessing. PreSto is a storage-centric preprocessing system leveraging In-Storage Processing (ISP), which offloads the bottlenecked preprocessing operations to our ISP units. We show that PreSto outperforms the baseline CPU-centric system with a 9.6× speedup in end-to-end preprocessing time, 4.3× enhancement in cost-efficiency, and 11.3× improvement in energy-efficiency on average for production-scale RecSys preprocessing.
Author Kim, Hyeseong
Lee, Yunjae
Rhu, Minsoo
Author_xml – sequence: 1
  givenname: Yunjae
  surname: Lee
  fullname: Lee, Yunjae
  email: yunjae408@kaist.ac.kr
  organization: KAIST,School of Electrical Engineering
– sequence: 2
  givenname: Hyeseong
  surname: Kim
  fullname: Kim, Hyeseong
  email: hyeseong.kim@kaist.ac.kr
  organization: KAIST,School of Electrical Engineering
– sequence: 3
  givenname: Minsoo
  surname: Rhu
  fullname: Rhu, Minsoo
  email: mrhu@kaist.ac.kr
  organization: KAIST,School of Electrical Engineering
BookMark eNotj81OwzAQhI0EElD6Bj34BRLWdmwn3KoCJVIRqC3namNvqkiNXTm59O0JP6cZzacZae7ZdYiBGFsIyIWA6rHerZa6AmtzCbLIAUCpKzavbFUqDUoaXYpbNh-GrgEDlVW21Hds-5loN8Ynvgy8DtlkEx6JP-OIfELnFB1NlXDku8swUs_bmPg-YRd-si252PcUPI5dDPw9ejoND-ymxdNA83-dsa_Xl_3qLdt8rOvVcpOh1OWYicaY1gkPhbGIVEjvfFEANMpodAgOwDelRgmqLWzbGIctUiWkk07JUqgZW_ztdkR0OKeux3Q5iN9zQqtvictRqg
CODEN IEEPAD
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1109/ISCA59077.2024.00033
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 Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
EISBN 9798350326581
EndPage 353
ExternalDocumentID 10609715
Genre orig-research
GrantInformation_xml – fundername: National Research Foundation
  funderid: 10.13039/501100001321
– fundername: SK Hynix
  funderid: 10.13039/100018058
– fundername: IC Design Education Center
  funderid: 10.13039/501100003836
GroupedDBID 6IE
6IH
ACM
ALMA_UNASSIGNED_HOLDINGS
CBEJK
RIE
RIO
ID FETCH-LOGICAL-a258t-1b66fc1d0467aae42dcd4400b365aca0c00db85a203f47fb6cafae912c2c32813
IEDL.DBID RIE
ISICitedReferencesCount 1
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001290320700023&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 02:35:15 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-a258t-1b66fc1d0467aae42dcd4400b365aca0c00db85a203f47fb6cafae912c2c32813
PageCount 14
ParticipantIDs ieee_primary_10609715
PublicationCentury 2000
PublicationDate 2024-June-29
PublicationDateYYYYMMDD 2024-06-29
PublicationDate_xml – month: 06
  year: 2024
  text: 2024-June-29
  day: 29
PublicationDecade 2020
PublicationTitle 2024 ACM/IEEE 51st Annual International Symposium on Computer Architecture (ISCA)
PublicationTitleAbbrev ISCA
PublicationYear 2024
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssib060973785
Score 2.3007753
Snippet Training recommendation systems (RecSys) faces several challenges as it requires the "data preprocessing" stage to preprocess an ample amount of raw data and...
SourceID ieee
SourceType Publisher
StartPage 340
SubjectTerms computational storage device
Costs
Data preprocessing
Graphics processing units
near data processing
neural network
Parallel processing
Power demand
Recommendation system
Throughput
Training
Title PreSto: An In-Storage Data Preprocessing System for Training Recommendation Models
URI https://ieeexplore.ieee.org/document/10609715
WOSCitedRecordID wos001290320700023&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/eLvHCXMwlV1NSwMxEA22ePCkYsVvcvC6mk02m8RbqRZ7KcVW6K1MJlkQdFu2W3-_yW6rXjx4C8lAYPI5ybz3CLktnDIpokiUkfHpBrIEpMNE2yIT6NCyzDViE2o81vO5mWzB6g0WxnvfJJ_5u1hs_vLdEjfxqSys8DxSHskO6SilWrDWbvLEFqG03MLjUmbuR9NBX4bgT4UwkEeSbBblcX-JqDRnyPDwn70fkd4PGo9Ovs-ZY7LnyxPyMqn8tF4-0H5JR2USilXYGegj1BCM_arN_w_2tOUkp-FySmdbPQgag86P0F0rqESjINr7ukdeh0-zwXOy1UdIgEtdJ6nN8wJTF0JcBeAz7tBlYU1akUtAYMiYs1oCZ6LIVGFzhAK8STlyFFyn4pR0y2XpzwjVqJVjhTVZxMqCMICRnwAdA2sl8-ekFx2yWLUUGIudLy7-qL8kB9HnMaeKmyvSrauNvyb7-Fm_raubZuC-ANKmmv0
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEA5aBT2pWPFtDl5Xs3lsNt5KtbRYS7EVvJVkkgVBt2W79feb7G7ViwdvIQkMTB6TSfJ9H0LXmZUqBmCRVCJc3WgeaWEhSk3GGVgwhNtKbEKORunrqxo3YPUKC-Ocqz6fuZtQrN7y7RxW4arMr_AkUB6JTbQlOKdxDddaT5_QxmQqGoBcTNTtYNLtCJ_-SZ8I0kCTTYJA7i8ZlSqK9Pb-aX8ftX_weHj8HWkO0IbLD9HzuHCTcn6HOzke5JEvFn5vwPe61L6zW9QIAN8f16zk2B9P8bRRhMAh7fzw5mpJJRwk0d6XbfTSe5h2-1GjkBBpKtIyik2SZBBbn-RKrR2nFiz3q9KwRGjQBAixJhWaEpZxmZkEdKadiilQYDSN2RFq5fPcHSOcQiotyYziAS2rmdIQGArAEm2MIO4EtYNDZouaBGO29sXpH_VXaKc_fRrOhoPR4xnaDf4PP6yoOketsli5C7QNn-XbsrisBvELEnqeRA
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=2024+ACM%2FIEEE+51st+Annual+International+Symposium+on+Computer+Architecture+%28ISCA%29&rft.atitle=PreSto%3A+An+In-Storage+Data+Preprocessing+System+for+Training+Recommendation+Models&rft.au=Lee%2C+Yunjae&rft.au=Kim%2C+Hyeseong&rft.au=Rhu%2C+Minsoo&rft.date=2024-06-29&rft.pub=IEEE&rft.spage=340&rft.epage=353&rft_id=info:doi/10.1109%2FISCA59077.2024.00033&rft.externalDocID=10609715