Deployment and verification of machine learning tool-chain based on kubernetes distributed clusters This paper is submitted for possible publication in the special issue on high performance distributed computing

In the field of software engineering, the environmental dependency conflict is a significant problem facing software engineers. Containerization (Pahl 2015 ) was proposed to resolve environmental dependency conflicts, Currently widely used in cloud computing and distributed systems. Simultaneously,...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:CCF transactions on high performance computing (Online) Ročník 3; číslo 2; s. 157 - 170
Hlavní autori: Cai, Haoyu, Wang, Chao, Zhou, Xuehai
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Singapore Springer Singapore 01.06.2021
Predmet:
ISSN:2524-4922, 2524-4930
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract In the field of software engineering, the environmental dependency conflict is a significant problem facing software engineers. Containerization (Pahl 2015 ) was proposed to resolve environmental dependency conflicts, Currently widely used in cloud computing and distributed systems. Simultaneously, in the process of large-scale application development and deployment, the microservice (Nadareishvili et al. 2016 ) architecture has the advantages of robust scalability and low coupling. Therefore, it is becoming increasingly favored by software developers. For example, Google is one of the few companies that need to manage the deployment and development of a large number of service components on hundreds of thousands of servers. With the concept of containerization and microservices at the core, an open-source distributed container management system called Kubernetes was developed. Kubernetes can not only maintain complete independence of applications but also improve the utilization of hardware resources, so it is affected by Internet companies and widely used by many institutions. In recent years, the demand for computing resources for machine learning-related applications is increasing, and the stand-alone computing for machine learning tasks is often unsustainable. Many data scientists will rely on distributed systems to provide sufficient computing resources for machine learning tasks. Kubernetes can be used for machine learning related applications and provides support for deployment on distributed systems. Meanwhile, it has many excellent features such as containerization and microservices. Therefore, the idea of developing and deploying machine-learning applications based on Kubernetes is favored by data scientists. Google has developed Kubeflow, a machine learning tool suite based on the Kubernetes system. Kubeflow can help data scientists run machine learning workloads on distributed clusters. For historical reasons, Kubeflow’s support for Tensorflow is quite complete, but the support of the framework is not perfect for Pytorch. Besides, although the pipeline component included in Kubeflow can build an entire machine learning workflow, this component is still dependent on the Google Cloud Platform. Therefore, Kubeflow pipeline is not friendly enough for developers who have no conditions to lease Google Cloud Service. This paper designs a complete solution to the problem of Kubeflow pipeline, and verifies the feasibility of the solution through an example, so that Kubernetes no longer depends on Google Cloud Service and better supports machine learning applications based on Pytorch. This solution ensures that data scientists can accomplish Pytorch-based deep learning applications on Kubernetes-based distributed systems development, demonstration, construction workflow, deployment services, operation, maintenance, and other functions. In principle, any data scientist who wants to use Pytorch to develop a machine-learning project can deploy their machine “learning applications on distributed systems by following such a set of solutions. The proposed approach makes the application stably and continuously running and reasonably scheduled on the cluster containing heterogeneous computing resources.
AbstractList In the field of software engineering, the environmental dependency conflict is a significant problem facing software engineers. Containerization (Pahl 2015 ) was proposed to resolve environmental dependency conflicts, Currently widely used in cloud computing and distributed systems. Simultaneously, in the process of large-scale application development and deployment, the microservice (Nadareishvili et al. 2016 ) architecture has the advantages of robust scalability and low coupling. Therefore, it is becoming increasingly favored by software developers. For example, Google is one of the few companies that need to manage the deployment and development of a large number of service components on hundreds of thousands of servers. With the concept of containerization and microservices at the core, an open-source distributed container management system called Kubernetes was developed. Kubernetes can not only maintain complete independence of applications but also improve the utilization of hardware resources, so it is affected by Internet companies and widely used by many institutions. In recent years, the demand for computing resources for machine learning-related applications is increasing, and the stand-alone computing for machine learning tasks is often unsustainable. Many data scientists will rely on distributed systems to provide sufficient computing resources for machine learning tasks. Kubernetes can be used for machine learning related applications and provides support for deployment on distributed systems. Meanwhile, it has many excellent features such as containerization and microservices. Therefore, the idea of developing and deploying machine-learning applications based on Kubernetes is favored by data scientists. Google has developed Kubeflow, a machine learning tool suite based on the Kubernetes system. Kubeflow can help data scientists run machine learning workloads on distributed clusters. For historical reasons, Kubeflow’s support for Tensorflow is quite complete, but the support of the framework is not perfect for Pytorch. Besides, although the pipeline component included in Kubeflow can build an entire machine learning workflow, this component is still dependent on the Google Cloud Platform. Therefore, Kubeflow pipeline is not friendly enough for developers who have no conditions to lease Google Cloud Service. This paper designs a complete solution to the problem of Kubeflow pipeline, and verifies the feasibility of the solution through an example, so that Kubernetes no longer depends on Google Cloud Service and better supports machine learning applications based on Pytorch. This solution ensures that data scientists can accomplish Pytorch-based deep learning applications on Kubernetes-based distributed systems development, demonstration, construction workflow, deployment services, operation, maintenance, and other functions. In principle, any data scientist who wants to use Pytorch to develop a machine-learning project can deploy their machine “learning applications on distributed systems by following such a set of solutions. The proposed approach makes the application stably and continuously running and reasonably scheduled on the cluster containing heterogeneous computing resources.
Author Cai, Haoyu
Wang, Chao
Zhou, Xuehai
Author_xml – sequence: 1
  givenname: Haoyu
  orcidid: 0000-0002-5010-3396
  surname: Cai
  fullname: Cai, Haoyu
  email: caihaoyu@mail.ustc.edu.cn
  organization: USTC
– sequence: 2
  givenname: Chao
  surname: Wang
  fullname: Wang, Chao
  organization: USTC
– sequence: 3
  givenname: Xuehai
  surname: Zhou
  fullname: Zhou, Xuehai
  organization: USTC
BookMark eNp9kM1OQyEQRompibX2BVzxAihwobRLU3-TJm50TbgwtNRbaOBem769aI1LVzPJnG_y5VyiUUwRELpm9IZRqm6L4JIJQjkjlNKZJIczNOaSCyIWDR397ZxfoGkp2wpxxSjnszGy97Dv0nEHsccmOvwJOfhgTR9SxMnjnbGbEAF3YHIMcY37lDpiNyZE3JoCDlfuY2ghR-ihYBdKn0M79PViu6H0kMsVOvemKzD9nRP0_vjwtnwmq9enl-Xdilgm1YE0zKu5MvNZwxlt5aIF2RonnWAg5lYtBECt7b1yDVXSNV6YuTXAqZfCGdY2E8RPf21OpWTwep_DzuSjZlR_m9InU7qa0j-m9KGGmlOoVDiuIettGnKsPf9LfQHYl3B0
Cites_doi 10.1109/MCC.2014.51
10.1007/978-3-540-30218-6_19
10.1007/978-3-319-33383-0_5
10.1109/COMPSAC.2018.00026
10.1109/MCC.2015.51
10.1007/978-1-4842-4470-8_46
10.3139/9783446456020
10.1145/2966884.2966912
10.1109/TSE.1987.232562
10.1007/978-1-4842-2766-4_12
10.1109/CVPR.2019.00075
10.1109/MS.2016.64
10.1007/978-1-4842-4470-8
ContentType Journal Article
Copyright China Computer Federation (CCF) 2021
Copyright_xml – notice: China Computer Federation (CCF) 2021
DBID AAYXX
CITATION
DOI 10.1007/s42514-021-00065-w
DatabaseName CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EISSN 2524-4930
EndPage 170
ExternalDocumentID 10_1007_s42514_021_00065_w
GroupedDBID -EM
0R~
406
AACDK
AAHNG
AAJBT
AASML
AATNV
AAUYE
ABAKF
ABDZT
ABECU
ABFTV
ABJNI
ABKCH
ABMQK
ABTEG
ABTKH
ABTMW
ABXPI
ACAOD
ACDTI
ACHSB
ACMLO
ACOKC
ACPIV
ACZOJ
ADKNI
ADTPH
ADURQ
ADYFF
AEFQL
AEJRE
AEMSY
AESKC
AFBBN
AFKRA
AFQWF
AGDGC
AGJBK
AGMZJ
AGQEE
AGRTI
AIGIU
AILAN
AITGF
AJZVZ
ALMA_UNASSIGNED_HOLDINGS
AMKLP
AMXSW
AMYLF
ARAPS
AXYYD
BENPR
BGLVJ
BGNMA
CCPQU
DPUIP
EBLON
EBS
EJD
FIGPU
FINBP
FNLPD
FSGXE
GGCAI
H13
HCIFZ
IKXTQ
IWAJR
J-C
JZLTJ
K7-
KOV
LLZTM
M4Y
NPVJJ
NQJWS
NU0
PT4
ROL
RSV
SJYHP
SNE
SNPRN
SOHCF
SOJ
SRMVM
SSLCW
STPWE
TSG
UOJIU
UTJUX
VEKWB
VFIZW
ZMTXR
AAYXX
ABBRH
ABDBE
ABFSG
ABRTQ
ACSTC
AEZWR
AFDZB
AFFHD
AFHIU
AFOHR
AHPBZ
AHWEU
AIXLP
ATHPR
AYFIA
CITATION
PHGZM
PHGZT
PQGLB
ID FETCH-LOGICAL-c157w-31f787a863210b59be5bad5d41e48c794ee271ff7d3075d3f4a8cae20f54da1b3
IEDL.DBID RSV
ISSN 2524-4922
IngestDate Sat Nov 29 04:01:15 EST 2025
Fri Feb 21 02:47:47 EST 2025
IsPeerReviewed true
IsScholarly true
Issue 2
Keywords Kubeflow
Pytorch
pipeline
Containerization
Kubernetes
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c157w-31f787a863210b59be5bad5d41e48c794ee271ff7d3075d3f4a8cae20f54da1b3
ORCID 0000-0002-5010-3396
PageCount 14
ParticipantIDs crossref_primary_10_1007_s42514_021_00065_w
springer_journals_10_1007_s42514_021_00065_w
PublicationCentury 2000
PublicationDate 20210600
2021-06-00
PublicationDateYYYYMMDD 2021-06-01
PublicationDate_xml – month: 6
  year: 2021
  text: 20210600
PublicationDecade 2020
PublicationPlace Singapore
PublicationPlace_xml – name: Singapore
PublicationTitle CCF transactions on high performance computing (Online)
PublicationTitleAbbrev CCF Trans. HPC
PublicationYear 2021
Publisher Springer Singapore
Publisher_xml – name: Springer Singapore
References Vitux.com. Install nfs server and client on ubuntu 18.04 lts[EB/OL] . (2020). http://vitux.com/installNFSserverandclientonubuntu
Awan, A.A., Hamidouche, K., Venkatesh, A., et al.: Efficient large message broadcast using NCCL and CUDA-aware MPI for deep learning. Proceedings of the 23rd European MPI Users’ Group Meeting. pp. 5–22 (2016)
MerkelDDocker: lightweight linux containers for consistent development and deploymentLinux J.201420142392
Nadareishvili, I., Mitra, R., McLarty, M., et al.: Microservice architecture: aligning principles, practices, and culture[M]. O’Reilly Media, Inc. (2016)
BalalaieAHeydarnooriAJamshidiPMicroservices architecture enables devops: migration to a cloud-native architectureIEEE Softw.2016333425210.1109/MS.2016.64
Mané D.: TensorBoard: TensorFlow’s visualization toolkit (2015)
Rezatofighi, H. et al.: Generalized intersection over union: A metric and a loss for bounding box regression. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. (2019)
BernsteinDContainers and cloud: from lxc to docker to kubernetesIEEE Cloud Comput201413818410.1109/MCC.2014.51
Redmon, J., Farhadi, A.: Yolov3: An incremental improvement. arXiv:1804.02767 (2018)
Luksa, M.: Kubernetes in action. Manning Publications (2017)
Hindman, B., Konwinski, A., Zaharia, M., et al.: Mesos: a platform for finegrained resource sharing in the data center. NSDI: volume 11:22–22 (2011)
Bisong, E.: Kubeflow and Kubeflow pipelines. Building Machine Learning and Deep Learning Models on Google Cloud Platform. Springer, New York, pp. 671–685 (2019)
PahlCContainerization and the paas cloudIEEE Cloud Comput.201523243110.1109/MCC.2015.51
Sandberg, R., Goldberg, D., Kleiman, S., et al.: Design and implementation of the sun network filesystem. In: Proceedings of the Summer USENIX conference, pp. 119–130 (1985)
BisongEKubeflow and Kubeflow Pipelines[M]//Building Machine Learning and Deep Learning Models on Google Cloud Platform2019BerkeleyApress67168510.1007/978-1-4842-4470-8
Chen, T., Li, M., Li, Y., et al.: Mxnet: a flexible and efficient machine learning library for heterogeneous distributed systems. (2015). arXiv:1512.01274
KooRTouegSCheckpointing and rollback-recovery for distributed systemsIEEE Trans. Softw. Eng.19871233110.1109/TSE.1987.232562
Kramer, O.: Scikitlearn. Machine learning for evolution strategies. Springer, New York, pp. 45–53 (2016)
Abadi, M., Barham, P., Chen, J., et al.: Tensorflow: a system for largescale machine learning. In: 12thUSENIX Symposium on Operating Systems Design and Implementation (OSDI16), pp. 265–283 (2016)
Ketkar, N.:Introduction to pytorch. Deep learning with python. Springer, New York, pp. 95–208 (2017)
Gabriel, E., Fagg, G.E., Bosilca, G., et al.: Open MPI: Goals, concept, and design of a next generation MPI implementation. European Parallel Virtual Machine/Message Passing Interface Users’ Group Meeting, pp. 97–104. Springer, Berlin (2004)
Zhang, Y., Yin, G., Wang, T., et al.: An insight into the impact of dockerfile evolutionary trajectories on quality and latency[C]. In: 2018 IEEE 42nd Annual Computer Software and Applications Conference (COMPSAC). IEEE 1, 138–143 (2018)
Kluyver, T., Ragankelley, B., PÉREZ, F., et al.: Jupyter notebooks a publishing format for reproducible computational workflows. ELPUB, pp. 87–90 (2016)
65_CR15
A Balalaie (65_CR3) 2016; 33
65_CR14
C Pahl (65_CR18) 2015; 2
65_CR13
D Merkel (65_CR16) 2014; 2014
65_CR23
65_CR19
R Koo (65_CR12) 1987; 1
65_CR17
D Bernstein (65_CR4) 2014; 1
65_CR7
65_CR11
65_CR22
65_CR8
65_CR10
65_CR21
E Bisong (65_CR5) 2019
65_CR9
65_CR20
65_CR6
65_CR1
65_CR2
References_xml – reference: Kramer, O.: Scikitlearn. Machine learning for evolution strategies. Springer, New York, pp. 45–53 (2016)
– reference: MerkelDDocker: lightweight linux containers for consistent development and deploymentLinux J.201420142392
– reference: Rezatofighi, H. et al.: Generalized intersection over union: A metric and a loss for bounding box regression. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. (2019)
– reference: Zhang, Y., Yin, G., Wang, T., et al.: An insight into the impact of dockerfile evolutionary trajectories on quality and latency[C]. In: 2018 IEEE 42nd Annual Computer Software and Applications Conference (COMPSAC). IEEE 1, 138–143 (2018)
– reference: Nadareishvili, I., Mitra, R., McLarty, M., et al.: Microservice architecture: aligning principles, practices, and culture[M]. O’Reilly Media, Inc. (2016)
– reference: Abadi, M., Barham, P., Chen, J., et al.: Tensorflow: a system for largescale machine learning. In: 12thUSENIX Symposium on Operating Systems Design and Implementation (OSDI16), pp. 265–283 (2016)
– reference: Gabriel, E., Fagg, G.E., Bosilca, G., et al.: Open MPI: Goals, concept, and design of a next generation MPI implementation. European Parallel Virtual Machine/Message Passing Interface Users’ Group Meeting, pp. 97–104. Springer, Berlin (2004)
– reference: Sandberg, R., Goldberg, D., Kleiman, S., et al.: Design and implementation of the sun network filesystem. In: Proceedings of the Summer USENIX conference, pp. 119–130 (1985)
– reference: Ketkar, N.:Introduction to pytorch. Deep learning with python. Springer, New York, pp. 95–208 (2017)
– reference: Luksa, M.: Kubernetes in action. Manning Publications (2017)
– reference: BernsteinDContainers and cloud: from lxc to docker to kubernetesIEEE Cloud Comput201413818410.1109/MCC.2014.51
– reference: BisongEKubeflow and Kubeflow Pipelines[M]//Building Machine Learning and Deep Learning Models on Google Cloud Platform2019BerkeleyApress67168510.1007/978-1-4842-4470-8
– reference: KooRTouegSCheckpointing and rollback-recovery for distributed systemsIEEE Trans. Softw. Eng.19871233110.1109/TSE.1987.232562
– reference: Kluyver, T., Ragankelley, B., PÉREZ, F., et al.: Jupyter notebooks a publishing format for reproducible computational workflows. ELPUB, pp. 87–90 (2016)
– reference: Bisong, E.: Kubeflow and Kubeflow pipelines. Building Machine Learning and Deep Learning Models on Google Cloud Platform. Springer, New York, pp. 671–685 (2019)
– reference: Hindman, B., Konwinski, A., Zaharia, M., et al.: Mesos: a platform for finegrained resource sharing in the data center. NSDI: volume 11:22–22 (2011)
– reference: BalalaieAHeydarnooriAJamshidiPMicroservices architecture enables devops: migration to a cloud-native architectureIEEE Softw.2016333425210.1109/MS.2016.64
– reference: Mané D.: TensorBoard: TensorFlow’s visualization toolkit (2015)
– reference: Chen, T., Li, M., Li, Y., et al.: Mxnet: a flexible and efficient machine learning library for heterogeneous distributed systems. (2015). arXiv:1512.01274
– reference: PahlCContainerization and the paas cloudIEEE Cloud Comput.201523243110.1109/MCC.2015.51
– reference: Vitux.com. Install nfs server and client on ubuntu 18.04 lts[EB/OL] . (2020). http://vitux.com/installNFSserverandclientonubuntu/
– reference: Redmon, J., Farhadi, A.: Yolov3: An incremental improvement. arXiv:1804.02767 (2018)
– reference: Awan, A.A., Hamidouche, K., Venkatesh, A., et al.: Efficient large message broadcast using NCCL and CUDA-aware MPI for deep learning. Proceedings of the 23rd European MPI Users’ Group Meeting. pp. 5–22 (2016)
– ident: 65_CR22
– volume: 2014
  start-page: 2
  issue: 239
  year: 2014
  ident: 65_CR16
  publication-title: Linux J.
– ident: 65_CR21
– volume: 1
  start-page: 81
  issue: 3
  year: 2014
  ident: 65_CR4
  publication-title: IEEE Cloud Comput
  doi: 10.1109/MCC.2014.51
– ident: 65_CR19
– ident: 65_CR8
  doi: 10.1007/978-3-540-30218-6_19
– ident: 65_CR13
  doi: 10.1007/978-3-319-33383-0_5
– ident: 65_CR9
– ident: 65_CR23
  doi: 10.1109/COMPSAC.2018.00026
– volume: 2
  start-page: 24
  issue: 3
  year: 2015
  ident: 65_CR18
  publication-title: IEEE Cloud Comput.
  doi: 10.1109/MCC.2015.51
– ident: 65_CR7
– ident: 65_CR1
– ident: 65_CR11
– ident: 65_CR6
  doi: 10.1007/978-1-4842-4470-8_46
– ident: 65_CR14
  doi: 10.3139/9783446456020
– ident: 65_CR2
  doi: 10.1145/2966884.2966912
– volume: 1
  start-page: 23
  year: 1987
  ident: 65_CR12
  publication-title: IEEE Trans. Softw. Eng.
  doi: 10.1109/TSE.1987.232562
– ident: 65_CR17
– ident: 65_CR10
  doi: 10.1007/978-1-4842-2766-4_12
– ident: 65_CR15
– ident: 65_CR20
  doi: 10.1109/CVPR.2019.00075
– volume: 33
  start-page: 42
  issue: 3
  year: 2016
  ident: 65_CR3
  publication-title: IEEE Softw.
  doi: 10.1109/MS.2016.64
– start-page: 671
  volume-title: Kubeflow and Kubeflow Pipelines[M]//Building Machine Learning and Deep Learning Models on Google Cloud Platform
  year: 2019
  ident: 65_CR5
  doi: 10.1007/978-1-4842-4470-8
SSID ssj0002710226
ssib053822361
Score 2.1817648
Snippet In the field of software engineering, the environmental dependency conflict is a significant problem facing software engineers. Containerization (Pahl 2015 )...
SourceID crossref
springer
SourceType Index Database
Publisher
StartPage 157
SubjectTerms 3. High Performance Distributed Computing
Computer Hardware
Computer Science
Computer Systems Organization and Communication Networks
Regular Paper
Subtitle This paper is submitted for possible publication in the special issue on high performance distributed computing
Title Deployment and verification of machine learning tool-chain based on kubernetes distributed clusters
URI https://link.springer.com/article/10.1007/s42514-021-00065-w
Volume 3
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVPQU
  databaseName: Computer Science Database
  customDbUrl:
  eissn: 2524-4930
  dateEnd: 20241212
  omitProxy: false
  ssIdentifier: ssj0002710226
  issn: 2524-4922
  databaseCode: K7-
  dateStart: 20190501
  isFulltext: true
  titleUrlDefault: http://search.proquest.com/compscijour
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: ProQuest Central
  customDbUrl:
  eissn: 2524-4930
  dateEnd: 20241212
  omitProxy: false
  ssIdentifier: ssj0002710226
  issn: 2524-4922
  databaseCode: BENPR
  dateStart: 20190501
  isFulltext: true
  titleUrlDefault: https://www.proquest.com/central
  providerName: ProQuest
– providerCode: PRVAVX
  databaseName: SpringerLINK Contemporary 1997-Present
  customDbUrl:
  eissn: 2524-4930
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0002710226
  issn: 2524-4922
  databaseCode: RSV
  dateStart: 20190501
  isFulltext: true
  titleUrlDefault: https://link.springer.com/search?facet-content-type=%22Journal%22
  providerName: Springer Nature
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV3JTsMwEB1By4ELZRVlqXzgBpaIa9fJkaUVp6oqi3qLvEJFSVDT0t_HdpNKlRASnDOKoueZebE9Mw_gQjFJbSQ62FCZYKoJx9JGDFPBTRRLxhi3QWyC9_vxaJQMyqawoqp2r64kQ6ZeNbs574oo9iUFgTjxYhPqju5iL9gwfHypvMhFMKkmioR8TDyJBt01wgjFNCGk7J75-bXrDLV-PRpYp9f43_fuwk75l4lulm6xBxsm24dGpeCAyoA-AHVvvOCvPyFEItPI-bUvHQqrhXKLPkKppUGltsQrmuX5BKs3Mc6Q5z-NnN37XJpp5g9wkfZjeL2ClnuiJnM_hKE4hOde9-nuAZeyC1hFjC9cVrYuikXc8e09kiXSMCk00zQyNFYufo1xkFrLtcsPTLctFbEShlxbRrWIZPsIalmemWNAtKO4UDS2RLptFyPS7Te1UKItleHC0iZcVlCnn8vpGulqjnLAL3X4pQG_dNGEqwrptIy04hfzk7-Zn8I2CYvlT1jOoDabzs05bKmv2biYtqB-2-0Phq3gat9Q7c5Y
linkProvider Springer Nature
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1bS8MwFD7oFPTFecV5zYNvGrBZsrSP4oWJc4hO2VvJVYezk3Vzf98kawcDEfS5h1K-nEuTnPN9ACeKSWoj0cCGygRTTTiWNmKYCm6iWDLGuA1iE7zdjrvd5KEYCsvLbvfySjJk6tmwm_OuiGLfUhAKJ54swhJ1Fcsz5j8-vZRe5CKYlIwiIR8TX0SD7hphhGKaEFJMz_z82vkKNX89GqrOTfV_37sOa8VfJrqYusUGLJhsE6qlggMqAnoL1JXxgr_-hBCJTCPn1751KKwWGlj0EVotDSq0JV7RaDDoY_Umehny9U8jZ_c-lmaY-QNcpD0Nr1fQck9Uf-xJGPJteL657lw2cSG7gFXE-MRlZeuiWMQNP94jWSINk0IzTSNDY-Xi1xgHqbVcu_zAdN1SESthyLllVItI1negkg0yswuINhQXisaWSLftYkS6_aYWStSlMlxYWoPTEur0c8qukc54lAN-qcMvDfilkxqclUinRaTlv5jv_c38GFaanftW2rpt3-3DKgkL509bDqAyGo7NISyrr1EvHx4Fd_sGh6HPog
linkToPdf http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1LS8NAEB60inixPrE-9-BNl5rtbpMcxVoUpRR80FvYpxZrUtrU_n13t0mxIIJ4zhDC7DwyszPfB3AmmaAm4E2sqYgxVSTEwgQMUx7qIBKMsdB4somw04l6vbj7bYvfT7uXV5KznQaH0pTm9aEy9fnim7W0gGI3XuCTKJ4uwwp1g_SuXn98KS3KejMp0UV8bCYuoXoONsIIxTQmpNik-fm1i9lq8arUZ6B29f_fvgkbxd8nupqZyxYs6XQbqiWzAyocfQdkSzsiYNc5RDxVyNq7Gynyp4gygz78CKZGBefEK8qzbIDlG--nyOVFhazc-0ToUeoau0g5eF7HrGWfyMHEgTOMd-G5ffN0fYsLOgYsAxZObbQ21rt51HRrP4LFQjPBFVM00DSS1q-1tuo1JlQ2bjDVMJRHkmtyaRhVPBCNPaikWar3AdGmDLmkkSHClmOMCFuHKi55Q0gdckNrcF6qPRnOUDeSOb6y119i9Zd4_SXTGlyUWk8KDxz_In7wN_FTWOu22snDXef-ENaJPzfXhDmCSj6a6GNYlZ95fzw68Zb3BegY2IY
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=Deployment+and+verification+of+machine+learning+tool-chain+based+on+kubernetes+distributed+clusters&rft.jtitle=CCF+transactions+on+high+performance+computing+%28Online%29&rft.au=Cai%2C+Haoyu&rft.au=Wang%2C+Chao&rft.au=Zhou%2C+Xuehai&rft.date=2021-06-01&rft.issn=2524-4922&rft.eissn=2524-4930&rft.volume=3&rft.issue=2&rft.spage=157&rft.epage=170&rft_id=info:doi/10.1007%2Fs42514-021-00065-w&rft.externalDBID=n%2Fa&rft.externalDocID=10_1007_s42514_021_00065_w
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2524-4922&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2524-4922&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2524-4922&client=summon