A Cloud-Based Framework for Machine Learning Workloads and Applications
Open accessAbstract
In this paper we propose a distributed architecture to provide machine learning practitioners with a set of tools and cloud services that cover the whole machine learning development cycle: ranging from the models creation, training, validation and testing to the models serving as a service, sharing and publication. In such respect, the DEEP-Hybrid-DataCloud framework allows transparent access to existing e-Infrastructures, effectively exploiting distributed resources for the most compute-intensive tasks coming from the machine learning development cycle. Moreover, it provides scientists with a set of Cloud-oriented services to make their models publicly available, by adopting a serverless architecture and a DevOps approach, allowing an easy share, publish and deploy of the developed models.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Packaging models in containers is a commonly reported method of integration, in which the model is accessible through REST or RPC interfaces.
“Packaging models in containers is a commonly reported method of integration, in which the model is accessible through representation state transfer (REST) or remote procedure call (RPC) interfaces [12,16,28,45].”
From Operationalizing machine learning models: a systematic literature review (2022) · cites this paper as [16]
Model warmup can be achieved either through a method provided by the ML framework or through issuing an empty query against the model.
“Model warmup can be achieved either through a method provided by the ML framework [28] or more generally through issuing an empty query against the model [16].”
From Operationalizing machine learning models: a systematic literature review (2022) · cites this paper as [16]
Models are commonly made available for serving predictions through a REST or RPC API.
“Models are commonly made available for serving predictions through a REST [12,16,26,29,40,45] or RPC [12,28,45] API.”
From Operationalizing machine learning models: a systematic literature review (2022) · cites this paper as [16]
Models packaged in containers are simply run directly as standalone services.
“Models packaged in containers are simply run directly as standalone services [16,19,28,29,45].”
From Operationalizing machine learning models: a systematic literature review (2022) · cites this paper as [16]
Using a consistent standard for APIs, such as OpenAPI, across models can facilitate system integration.
“When deploying numerous different models, using a consistent standard for application programming interfaces (APIs), such as OpenAPIfoot_1 , across models can facilitate system integration [16].”
From Operationalizing machine learning models: a systematic literature review (2022) · cites this paper as [16]
Serving-related challenges are some of the ones most often reported in the literature, usually from a performance perspective.
“Serving-related challenges are some of the ones most often reported in the literature [5-7, 9, 11, 12, 16, 20, 28, 34, 44, 53, 54], usually from a performance perspective.”
From Operationalizing machine learning models: a systematic literature review (2022) · cites this paper as [16]
Models packaged in containers are typically accessed through either a REST API or an RPC interface.
“Models packaged in containers are typically accessed through either a REST API [16,40], such as Flask 18 [20], or RPC interface [12,45], such as Apache Thrift 19 [28].”
From Operationalizing machine learning models: a systematic literature review (2022) · cites this paper as [16]
Having multiple containers deployed often necessitates a container orchestration tool such as Kubernetes or Apache Mesos.
“Having multiple containers deployed will often necessitate the use of a container orchestration tool such as Kubernetes 20 [29,40,43,44] or Apache Mesos 21 [16].”
From Operationalizing machine learning models: a systematic literature review (2022) · cites this paper as [16]
Apache Kafka is a commonly reported messaging and communications framework, used for logging predictions, notifying model hosts of updates, and general inter-container communication.
“A commonly reported messaging and communications framework is Apache Kafka 28 , used for logging predictions [28], notifying model hosts of updates [16,29] and general inter-container communication [44].”
From Operationalizing machine learning models: a systematic literature review (2022) · cites this paper as [16]
Serverless frameworks like Apache OpenWhisk and Knative may be used to deploy models as functions and provide horizontal scaling, both building on top of Kubernetes.
“Serverless frameworks like Apache OpenWhisk 25 and Knativefoot_11 may be used to deploy models as functions and provide horizontal scaling [16,43], both building on top of Kubernetes.”
From Operationalizing machine learning models: a systematic literature review (2022) · cites this paper as [16]
Docker is by far the most frequently reported solution for packaging ML models in containers.
“Docker 17 is by far the most frequently reported solution [12,16,20,26,28,40,44].”
From Operationalizing machine learning models: a systematic literature review (2022) · cites this paper as [16]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126 · see citing works
MLlib: Machine Learning in Apache SparkPreprint
Meng, X., Bradley, J. K., Yavuz, B., Sparks, E., Venkataraman, S., Liu, D., Freeman, J., Tsai, D., Amde, M., Owen, S., et al. (2015) · Journal of machine learning research · doi:10.48550/arxiv.1505.06807 · Cited by 1,898 · see citing works
Scaling Distributed Machine Learning with the Parameter Server
Li, M. (2014) · International Conference on Big Data Science and Computing (BigDataScience) · doi:10.1145/2640087.2644155 · Cited by 1,000 · see citing works
Machine Learning and Deep Learning frameworks and libraries for large-scale data mining: a survey
Nguyen, G., Dlugolinský, Š., Bobák, M., Tran, V., García, Á. L., Heredia, I., Malík, P., & Hluchý, L. (2019) · Artificial Intelligence Review · doi:10.1007/s10462-018-09679-z · Cited by 924 · see citing works
Self-citation(2015) · neural information processing systems · Cited by 912 · see citing works
Science Teaching and Development Of Thinking
Lawson, A. E. (1994) · Cited by 506 · see citing works
Communication Efficient Distributed Machine Learning with the Parameter Server
Li, M., Andersen, D. G., Smola, A. J., & Yu, K. (2014) · neural information processing systems · Cited by 475 · see citing works
An imaging-in-flow system for automated analysis of marine microplankton
Sieracki, C. K., Sieracki, M. E., & Yentsch, C. S. (1998) · Marine Ecology Progress Series · doi:10.3354/meps168285 · Cited by 438 · see citing works
MLaaS: Machine Learning as a Service
Ribeiro, M., Grolinger, K., & Capretz, M. A. M. (2015) · IEEE International Conference on Machine Learning and Applications (ICMLA) · doi:10.1109/icmla.2015.152 · Cited by 395 · see citing works
MLbase: A Distributed Machine-learning System
Kraska, T., Talwalkar, A., & Duchi, J. C. (2013) · Cited by 310 · see citing works
Motivation for and Evaluation of the First Tensor Processing Unit
Jouppi, N. P., Young, C., Patil, N. P., & Patterson, D. A. (2018) · IEEE Micro · doi:10.1109/mm.2018.032271057 · Cited by 241 · see citing works
A machine learning model for improving healthcare services on cloud computing environment
Abdelaziz, A., Elhoseny, M., Salama, A. S., & Riad, A. M. (2018) · Measurement · doi:10.1016/j.measurement.2018.01.022 · Cited by 234 · see citing works
A data ecosystem to support machine learning in materials science
Blaiszik, B. J., Ward, L. T., Schwarting, M., et al. (2019) · MRS Communications · doi:10.1557/mrc.2019.118 · Cited by 179 · see citing works
Serving Deep Learning Models in a Serverless Platform
Ishakian, V., Muthusamy, V., & Slominski, A. (2018) · IEEE International Conference on Cloud Engineering · doi:10.1109/ic2e.2018.00052 · Cited by 169 · see citing works
The Kipoi repository accelerates community exchange and reuse of predictive models for genomics
Avsec, Ž., Kreuzhuber, R., Israeli, J., Xu, N., Cheng, J., Shrikumar, A., Banerjee, A., Kim, D. S., Beier, T., Urban, L., et al. (2019) · Nature Biotechnology · doi:10.1038/s41587-019-0140-0 · Cited by 148 · see citing works
Cloud-based deep learning of big EEG data for epileptic seizure prediction
Hosseini, M.-P., Hamid Soltanian-zadeh, A., Elisevich, K. V., & Pompili, D. (2016) · IEEE Global Conference on Signal and Information Processing · doi:10.1109/globalsip.2016.7906022 · Cited by 118 · see citing works
Chen, X., Dallmeier-Tiessen, S., Dasler, R., Feger, S. S., Fokianos, P., Gonzalez, J. B., Hirvonsalo, H., Kousidis, D., Lavasa, A., Mele, S., et al. (2018) · Nature Physics · doi:10.1038/s41567-018-0342-2 · Cited by 114 · see citing works
Towards Unified Data and Lifecycle Management for Deep Learning
Miao, H., Li, A., Davis, L. S., & Deshpande, A. (2017) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde.2017.112 · Cited by 104 · see citing works
Dynamic Management of Virtual Infrastructures
Caballer, M., Blanquer, I., Moltó, G., et al. (2014) · Journal of Grid Computing · doi:10.1007/s10723-014-9296-5 · Cited by 88 · see citing works
Self-citationApplication of Deep Learning to Sentiment Analysis for recommender system on cloud
Preethi, G., Krishna, P. V., Obaidat, M. S., et al. (2017) · International Conference on Computer, Information and Telecommunication Systems · doi:10.1109/cits.2017.8035341 · Cited by 88 · see citing works
Yao, Y., Xiao, Z., Wang, B., Viswanath, B., Zheng, H.-T., & Zhao, B. Y. (2017) · ACM Internet Measurement Conference (IMC) · doi:10.1145/3131365.3131372 · Cited by 86 · see citing works
Log analysis in cloud computing environment with Hadoop and Spark
Lin, X., Wang, P., & Wu, B. Y. (2013) · IEEE International Conference on Broadband Network & Multimedia Technology (IC-BNMT) · doi:10.1109/icbnmt.2013.6823956 · Cited by 78 · see citing works
K-means Clustering in the Cloud -- A Mahout Test
Esteves, R. M., Pais, R., & Rong, C. (2011) · doi:10.1109/waina.2011.136 · Cited by 77 · see citing works
INDIGO-DataCloud: a Platform to Facilitate Seamless Access to E-Infrastructures
Salomoni, D., Campos, I., Gaido, L., et al. (2018) · Journal of Grid Computing · doi:10.1007/s10723-018-9453-3 · Cited by 74 · see citing works
DLHub: Model and Data Serving for Science
Chard, R., Li, Z., Chard, K., Ward, L. T., Babuji, Y., Woodard, A. E., Tuecke, S. J., Blaiszik, B. J., Franklin, M. J., & Foster, I. T. (2019) · IEEE International Parallel and Distributed Processing Symposium (IPDPS) · doi:10.1109/ipdps.2019.00038 · Cited by 73 · see citing works
Granules: A lightweight, streaming runtime for cloud computing with support, for Map-Reduce
Pallickara, S., Ekanayake, J., & Fox, G. (2009) · doi:10.1109/clustr.2009.5289160 · Cited by 64 · see citing works
Enabling rootless Linux Containers in multi-user environments: The udocker tool
Gomes, J., Bagnaschi, E., Campos, I., David, M., Alves, L. M., Martins, J. P. F., Pina, J., García, Á. L., & Orviz, P. (2018) · Computer Physics Communications · doi:10.1016/j.cpc.2018.05.021 · Cited by 58 · see citing works
Self-citationA scalable machine learning online service for big data real-time analysis
Baldominos, A., Albacete, E., Sáez, Y., et al. (2014) · doi:10.1109/cibd.2014.7011537 · Cited by 56 · see citing works
JSON Schema: A Media Type for Describing JSON Documents
Andrews, H. R., & Wright, A. (2019) · Cited by 54 · see citing works
CloudSVM: Training an SVM Classifier in Cloud Computing Systems
Çatak, F. Ö., & Balaban, M. E. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-37015-1_6 · Cited by 43 · see citing works
The EGI Federated Cloud e-Infrastructure
Fernández, E., Scardaci, D., & García, Á. L. (2015) · Procedia Computer Science · doi:10.1016/j.procs.2015.09.235 · Cited by 39 · see citing works
Messina, F., Pappalardo, G. C., Rosaci, D., et al. (2015) · Future Generation Computer Systems · doi:10.1016/j.future.2015.07.013 · Cited by 38 · see citing works
Orchestrating Complex Application Architectures in Heterogeneous Clouds
Caballer, M., Zala, S., García, Á. L., et al. (2017) · Journal of Grid Computing · doi:10.1007/s10723-017-9418-y · Cited by 35 · see citing works
Self-citationChan, S., Stone, T., Szeto, K. P., et al. (2013) · Proceedings of the ... ACM International Conference on Information & Knowledge Management · doi:10.1145/2505515.2508198 · Cited by 28 · see citing works
Assessment of machine learning algorithms in cloud computing frameworks
Li, K., Gibson, C., Ho, D., et al. (2013) · IEEE Systems and Information Engineering Design Symposium (SIEDS) · doi:10.1109/sieds.2013.6549501 · Cited by 19 · see citing works
DEEPaaS API: a REST API for Machine Learning and Deep Learning models
García, Á. L. (2019) · The Journal of Open Source Software · doi:10.21105/joss.01517 · Cited by 15 · see citing works
Resource provisioning in Science Clouds: Requirements and challenges
García, Á. L., Fernández, E., Orviz, P., et al. (2017) · Software Practice and Experience · doi:10.1002/spe.2544 · Cited by 13 · see citing works
Data automation at light sources
Blaiszik, B. J., Chard, K., Chard, R., Foster, I. T., & Ward, L. T. (2019) · AIP conference proceedings · doi:10.1063/1.5084563 · Cited by 8 · see citing works
Convolutional Neural Networks for Phytoplankton identification and classification
Lloret, L., Heredia, I., Aguilar, F. J., et al. (2018) · Biodiversity Information Science and Standards · doi:10.3897/biss.2.25762 · Cited by 4 · see citing works
Deep learning for weed identification based on seed images
Pando, F., Heredia, I., Pérez, C. A., et al. (2018) · Biodiversity Information Science and Standards · doi:10.3897/biss.2.25749 · Cited by 2 · see citing works
Heredia, I., Iglesias, L. L., Vykozlov, V., et al. (2019) · DIGITAL.CSIC (Spanish National Research Council (CSIC)) · doi:10.20350/digitalcsic/9895 · Cited by 2 · see citing works
Heredia, I., Iglesias, L. L., & Vykozlov, V. (2018) · DIGITAL.CSIC (Spanish National Research Council (CSIC)) · doi:10.20350/digitalcsic/9897 · Cited by 1 · see citing works
Heredia, I., Lloret Iglesias, L., & Vykozlov, V. (2019) · DIGITAL.CSIC (Spanish National Research Council (CSIC)) · doi:10.20350/digitalcsic/9892 · Cited by 1 · see citing works
Self-citationHeredia, I., Lloret Iglesias, L., & Vykozlov, V. (2019) · DIGITAL.CSIC (Spanish National Research Council (CSIC)) · doi:10.20350/digitalcsic/9896 · Cited by 1 · see citing works
Self-citation
Flagged first, then by citations
Research works printed but not linked: 3
- L. E. Li, E. Chen, J. Hermann, P. Zhang, and L. Wang, ''Scaling machine learning as a service,'' in Proc. 3rd Int. Conf. Predictive Appl. (APIs), vol. 67, C. Hardgrove, L. Dorard, K. Thompson, and F. Douetteau, Eds. Boston, MA, USA: Microsoft NERD, Oct. 2017, pp. 14-29.
- C. Castellani and M. Edwards, Eds., Marine Plankton: A Practical Guide to Ecology, Methodology, and Taxonomy, vol. 1. London, U.K.: Oxford Univ. Press, 2017.
- K. Ueda, ''Inaturalist research-grade observations,'' Inaturalist. org. Occurrence Dataset, Tech. Rep., 2017. [Online]. Available: https://www. gbif.org/dataset/50c9509d-22c7-4a22-a47d-8c48425ef4a7, doi: 10.15468/ ab3s5x · DOI
Cited by
Sorted by citations
Machine Learning Operations (MLOps): Overview, Definition, and Architecture
Kreuzberger, D., Kühl, N., & Hirschl, S. (2023) · IEEE Access · doi:10.1109/access.2023.3262138 · Cited by 766 · see citing works
Steidl, M., Felderer, M., & Ramler, R. (2023) · Journal of Systems and Software · doi:10.1016/j.jss.2023.111615 · Cited by 126 · see citing works
Operationalizing machine learning models: a systematic literature review
Kolltveit, A. B., & Li, J. (2022) · Proceedings of the 1st Workshop on Software Engineering for Responsible AI · doi:10.1145/3526073.3527584 · Cited by 34 · see citing works
Building A Platform for Machine Learning Operations from Open Source Frameworks
Liu, Y., Ling, Z., Huo, B., Wang, B., Chen, T., & Mouine, E. (2020) · IFAC-PapersOnLine · doi:10.1016/j.ifacol.2021.04.161 · Cited by 32 · see citing works
- Authors
- Lopez Garcia, A., De Lucas, J. M., Antonacci, M., Zu Castell, W., David, M., Hardt, M., Lloret Iglesias, L., Molto, G., Plociennik, M., Tran, V., Alic, A. S., Caballer, M., Plasencia, I. C., Costantini, A., Dlugolinsky, S., Duma, D. C., Donvito, G., Gomes, J., Heredia Cacha, I., Ito, K., Kozlov, V. Y., Nguyen, G., Orviz Fernandez, P., Sustr, Z., & Wolniewicz, P.
- Affiliations
- Istituto Nazionale di Fisica Nucleare, Sezione di Bari (IT)
- LIP - Laboratory of Instrumentation and Experimental Particle Physics (PT)
- Karlsruhe Institute of Technology (DE)
- Slovak Academy of Sciences (SK)
- Instituto de Instrumentación para Imagen Molecular (ES)
- Istituto Nazionale di Fisica Nucleare, Centro Nazionale Analisi Fotogrammi (IT)
- Helmholtz Munich (DE)
- ORCID
- Lopez Garcia, Alvaro
- De Lucas, Jesus Marco
- Antonacci, Marica
- Zu Castell, Wolfgang
- David, Mario
- Hardt, Marcus
- Lloret Iglesias, Lara
- Molto, Germen
- Plociennik, Marcin
- Tran, Viet
- Alic, Andy S.
- Caballer, Miguel
- Plasencia, Isabel Campos
- Costantini, Alessandro
- Dlugolinsky, Stefan
- Duma, Doina Cristina
- Donvito, Giacinto
- Gomes, Jorge
- Heredia Cacha, Ignacio
- Ito, Keiichi
- Kozlov, Valentin Y.
- Nguyen, Giang
- Orviz Fernandez, Pablo
- Sustr, Zdenek
- Wolniewicz, Pawel
- Venue
- IEEE Access
- Published
- 2020
- Volume
- 8
- Pages
- 18681-18692
- ISSN
- 2169-3536 (ISSN-L)
- OpenAlex
- W2998827957
- Cited by
- 108
- References
- 44
- Citations by year
- 2026: 12 · 2025: 23 · 2024: 19 · 2023: 16 · 2022: 16 · 2021: 15 · 2020: 7
- FWCI
- 12.70
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- IEEE Access (publishedVersion, cc-by)
- Repository KITopen (Karlsruhe Institute of Technology) (submittedVersion, cc-by)
- Portuguese National Funding Agency for Science, Research and Technology (RCAAP Project by FCT) (submittedVersion, other-oa)
- DIGITAL.CSIC (Spanish National Research Council (CSIC)) (submittedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Site cant be reached (submittedVersion, cc-by)
- PuSH - Publication Server of Helmholtz Zentrum München (submittedVersion, cc-by)
- RiuNet (Universitat Politècnica de València) (submittedVersion, cc-by-nc-nd)
- KITopen (cc-by)
- Funders
- DEEP-Hybrid-DataCloud “Designing and Enabling E-infrastructures for intensive Processing in a Hybrid DataCloud”
- European Commission (777435)
- Topics
- Physical Sciences › Computer Science › Information Systems › Cloud Computing and Resource Management
- Physical Sciences › Computer Science › Computer Networks and Communications › Distributed and Parallel Computing Systems
- Social Sciences › Decision Sciences › Information Systems and Management › Scientific Computing and Data Management
- Keywords
- machine learning workloads, cloud computing, cyberinfrastructure, distributed computing, serverless architecture, DevOps, model serving, hybrid cloud
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
