A Cloud-Based Framework for Machine Learning Workloads and Applications

Alvaro Lopez Garcia, Jesus Marco De Lucas, Marica Antonacci, Wolfgang Zu Castell, Mario David, Marcus Hardt, Lara Lloret Iglesias, Germen Molto, Marcin Plociennik, Viet Tran, Andy S. Alic, Miguel Caballer, Isabel Campos Plasencia, Alessandro Costantini, Stefan Dlugolinsky, Doina Cristina Duma, Giacinto Donvito, Jorge Gomes, Ignacio Heredia Cacha, Keiichi Ito, Valentin Y. Kozlov, Giang Nguyen, Pablo Orviz Fernandez, Zdenek Sustr, and Pawel Wolniewicz. A Cloud-Based Framework for Machine Learning Workloads and Applications. IEEE Access, 8:18681–18692, 2020.

Download

[1.8MB pdf]  [HTML] 

Abstract

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.

BibTeX Entry

@article{DEEP-Access,
   abstract = {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.},
   author = {Alvaro Lopez Garcia and Jesus Marco De Lucas and Marica Antonacci and Wolfgang Zu Castell and Mario David and Marcus Hardt and Lara Lloret Iglesias and Germen Molto and Marcin Plociennik and Viet Tran and Andy S. Alic and Miguel Caballer and Isabel Campos Plasencia and Alessandro Costantini and Stefan Dlugolinsky and Doina Cristina Duma and Giacinto Donvito and Jorge Gomes and Ignacio Heredia Cacha and Keiichi Ito and Valentin Y. Kozlov and Giang Nguyen and Pablo Orviz Fernandez and Zdenek Sustr and Pawel Wolniewicz},
   doi = {10.1109/ACCESS.2020.2964386},
   issn = {2169-3536},
   journal = {IEEE Access},
   pages = {18681-18692},
   title = {A Cloud-Based Framework for Machine Learning Workloads and Applications},
   volume = {8},
   url = {https://ieeexplore.ieee.org/document/8950411/},
   year = {2020}
}