Accelerating Cloud-Native Databases with Distributed PMem Stores

Relational databases have gone through a phase of architectural transition from a monolithic to a distributed architecture to take full advantage of cloud technology. These distributed databases can leverage remote storage to maintain larger amounts of data than monolithic databases at the cost of i...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Data engineering s. 3043 - 3057
Hlavní autori: Sun, Jason, Ma, Haoxiang, Zhang, Li, Liu, Huicong, Shi, Haiyang, Luo, Shangyu, Wu, Kai, Bruhwiler, Kevin, Zhu, Cheng, Nie, Yuanyuan, Chen, Jianjun, Zhang, Lei, Liang, Yuming
Médium: Konferenčný príspevok..
Jazyk:English
Vydavateľské údaje: IEEE 01.04.2023
Predmet:
ISSN:2375-026X
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract Relational databases have gone through a phase of architectural transition from a monolithic to a distributed architecture to take full advantage of cloud technology. These distributed databases can leverage remote storage to maintain larger amounts of data than monolithic databases at the cost of increased latency. At ByteDance, we have built a distributed database called veDB based on the popular compute-storage separation architecture, however we have observed the system is unable to provide both low latency and high throughput required by some business critical applications, such as batched order processing.In this paper we present our novel approaches to tackle this problem. We have modified our system's storage to utilize persistent memory (PMem) coupled with a remote direct memory access (RDMA) network to reduce read/write latency and increase the throughput. We also propose a query push-down framework to push partial computations to the PMem storage layer to accelerate analytical queries and reduce the impact of the transaction workload in the computation layer. Our experiments show that our methods improve the throughput by up to 1.5× and reduce latency by up to 20× for standard benchmarks and real-world applications.
AbstractList Relational databases have gone through a phase of architectural transition from a monolithic to a distributed architecture to take full advantage of cloud technology. These distributed databases can leverage remote storage to maintain larger amounts of data than monolithic databases at the cost of increased latency. At ByteDance, we have built a distributed database called veDB based on the popular compute-storage separation architecture, however we have observed the system is unable to provide both low latency and high throughput required by some business critical applications, such as batched order processing.In this paper we present our novel approaches to tackle this problem. We have modified our system's storage to utilize persistent memory (PMem) coupled with a remote direct memory access (RDMA) network to reduce read/write latency and increase the throughput. We also propose a query push-down framework to push partial computations to the PMem storage layer to accelerate analytical queries and reduce the impact of the transaction workload in the computation layer. Our experiments show that our methods improve the throughput by up to 1.5× and reduce latency by up to 20× for standard benchmarks and real-world applications.
Author Wu, Kai
Bruhwiler, Kevin
Liang, Yuming
Zhang, Li
Zhu, Cheng
Luo, Shangyu
Zhang, Lei
Nie, Yuanyuan
Ma, Haoxiang
Liu, Huicong
Shi, Haiyang
Chen, Jianjun
Sun, Jason
Author_xml – sequence: 1
  givenname: Jason
  surname: Sun
  fullname: Sun, Jason
  email: jason.sun@bytedance.com
– sequence: 2
  givenname: Haoxiang
  surname: Ma
  fullname: Ma, Haoxiang
– sequence: 3
  givenname: Li
  surname: Zhang
  fullname: Zhang, Li
– sequence: 4
  givenname: Huicong
  surname: Liu
  fullname: Liu, Huicong
– sequence: 5
  givenname: Haiyang
  surname: Shi
  fullname: Shi, Haiyang
  organization: ByteDance Inc.,ByteDance US Infrastructure System Lab
– sequence: 6
  givenname: Shangyu
  surname: Luo
  fullname: Luo, Shangyu
  organization: ByteDance Inc.,ByteDance US Infrastructure System Lab
– sequence: 7
  givenname: Kai
  surname: Wu
  fullname: Wu, Kai
  email: wukai@microsoft.com
  organization: ByteDance Inc.,ByteDance US Infrastructure System Lab
– sequence: 8
  givenname: Kevin
  surname: Bruhwiler
  fullname: Bruhwiler, Kevin
  email: kevin.bruhwiler@gmail.com
– sequence: 9
  givenname: Cheng
  surname: Zhu
  fullname: Zhu, Cheng
– sequence: 10
  givenname: Yuanyuan
  surname: Nie
  fullname: Nie, Yuanyuan
– sequence: 11
  givenname: Jianjun
  surname: Chen
  fullname: Chen, Jianjun
  organization: ByteDance Inc.,ByteDance US Infrastructure System Lab
– sequence: 12
  givenname: Lei
  surname: Zhang
  fullname: Zhang, Lei
– sequence: 13
  givenname: Yuming
  surname: Liang
  fullname: Liang, Yuming
BookMark eNotjMtOwzAQRQ0CiVLyB13kB1L8iGPPjiopUKk8JLpgV02cCRilCYpdEH9PJLiLc3Q295Kd9UNPjC0EXwrB4XpTVmuttdBLyaVa8gnqhCVgwCrNlZTSwCmbSWV0xmXxesGSED74NMiF0HzGblbOUUcjRt-_pWU3HJvscYovSiuMWGOgkH77-J5WPsTR18dITfr8QIf0JQ4jhSt23mIXKPn3nO1u17vyPts-3W3K1TbzIGLmZFMAYGtqqyy2VnGTAymrUTgDEhpNrUMLtUZpisI6k-coDIJAa3nj1Jwt_m49Ee0_R3_A8WcvuLB5oUD9AuCiTBs
CODEN IEEPAD
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1109/ICDE55515.2023.00233
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan (POP) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
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
Discipline Computer Science
EISBN 9798350322279
EISSN 2375-026X
EndPage 3057
ExternalDocumentID 10184639
Genre orig-research
GroupedDBID 6IE
6IH
6IL
6IN
AAWTH
ABLEC
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IEGSK
IJVOP
OCL
RIE
RIL
RIO
ID FETCH-LOGICAL-i91t-c2d699af7b838af830749e385a1c7929d5efca89b5a27668c744a17a91a880dc3
IEDL.DBID RIE
IngestDate Wed Aug 27 02:07:22 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i91t-c2d699af7b838af830749e385a1c7929d5efca89b5a27668c744a17a91a880dc3
PageCount 15
ParticipantIDs ieee_primary_10184639
PublicationCentury 2000
PublicationDate 2023-April
PublicationDateYYYYMMDD 2023-04-01
PublicationDate_xml – month: 04
  year: 2023
  text: 2023-April
PublicationDecade 2020
PublicationTitle Data engineering
PublicationTitleAbbrev ICDE
PublicationYear 2023
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0000941150
Score 2.23651
Snippet Relational databases have gone through a phase of architectural transition from a monolithic to a distributed architecture to take full advantage of cloud...
SourceID ieee
SourceType Publisher
StartPage 3043
SubjectTerms Benchmark testing
cloud computing
cloud-native database
Computer architecture
distributed database
Distributed databases
PMem
Query processing
query push-down
RDMA
Relational databases
Technological innovation
Throughput
Title Accelerating Cloud-Native Databases with Distributed PMem Stores
URI https://ieeexplore.ieee.org/document/10184639
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEA62ePBUHxXf5OB12-4jj7kpfaCgpWCR3so0yUpBt9Ju_f1m0m314sHbEpZdmDB8M5N838fYbYY5oFZAFGXfoCCISGcoI5GCf1-ajgks19cnNRzqyQRGFVk9cGGcc-HymWvRYzjLtwuzplFZm9SlMg-pNVZTSm7IWruBiu9TqLqp6HFxB9qP3V5f-IpAtMgjnEYnZI_7y0QlYMig8c-_H7LmDxuPj3Y4c8T2XHHMGls7Bl5l5wm7uzfGgwhtafHGu--LtY2GQdeb97BEwqsVp7kr75FaLhldOctHz-6Dv_jO262abDzoj7sPUWWQEM0hLiOTWAmAuZrpVGOufbpm4FItMDbKlz1WuNyghpnAREmpjcoyjBVCjD5rrUlPWb1YFO6McSnQGgkmn7k8cyRZCjpB8N-XQqRJfM6aFJDp50YCY7qNxcUf65fsgGK-ueJyxerlcu2u2b75Kuer5U3YuG-x4piQ
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LT8JAEN4omugJHxjf7sFrgT72MTcNSCBCQyIx3Miw3RoSLQaKv9-dUtCLB2_NpmmT2Uy-mdn9vo-x-whTQK2AKMquQUEQno5QeiIE9740TVOwXF_7Ko71eAzDkqxecGGstcXlM1unx-IsP5mbFY3KGqQuFTlI3WV7ZJ3lr-la25GK61SovikJcn4TGr1W-0m4mkDUySWchidkkPvLRqVAkU71n_8_YrUfPh4fbpHmmO3Y7IRVN4YMvMzPU_bwaIyDEdrU7I233uerxIsLZW_exhwJsZacJq-8TXq5ZHVlEz4c2A_-4npvu6yxUedp1Op6pUWCNwM_90yQSABM1VSHGlPtEjYCG2qBvlGu8EmETQ1qmAoMlJTaqChCXyH46PI2MeEZq2TzzJ4zLgUmRoJJpzaNLImWgg4Q3PelEGHgX7AaBWTyuRbBmGxicfnH-h076I4G_Um_Fz9fsUOK__rCyzWr5IuVvWH75iufLRe3xSZ-A9Ngm9c
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=Data+engineering&rft.atitle=Accelerating+Cloud-Native+Databases+with+Distributed+PMem+Stores&rft.au=Sun%2C+Jason&rft.au=Ma%2C+Haoxiang&rft.au=Zhang%2C+Li&rft.au=Liu%2C+Huicong&rft.date=2023-04-01&rft.pub=IEEE&rft.eissn=2375-026X&rft.spage=3043&rft.epage=3057&rft_id=info:doi/10.1109%2FICDE55515.2023.00233&rft.externalDocID=10184639