Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 8
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Rescheduling serverless workloads across the cloud-to-edge continuum

  • Risco, Sebastián
  • Alarcón-Marín, Caterina
  • Langarita-Benítez, Sergio
  • Caballer Fernández, Miguel|||0000-0001-9393-3077
  • Moltó, Germán|||0000-0002-8049-253X
[EN] Serverless computing was a breakthrough in Cloud computing due to its high elasticity capabilities and fine-grained pay-per-use model offered by the main public Cloud providers. Meanwhile, open-source serverless platforms supporting the FaaS (Function as a Service) model allow users to take advantage of many of their benefits while operating on the on-premises platforms of organizations. This opens the possibility to deploy and exploit them on the different layers of the cloud-to-edge continuum, either on IoT (Internet of Things) devices located at the Edge (i.e. next to data acquisition devices), in on-premises clusters closer to the data sources (i.e. Fog computing) or directly on the Cloud. This paper presents two strategies to mitigate the overload that disparate data ingestion rates may cause in low-powered devices at the Edge or Fog layers. To this end, it is proposed to delegate and reschedule serverless jobs between the different layers of the cloud-to-edge continuum using an open-source platform for event-driven file processing. To demonstrate the performance of these strategies, a use case for fire detection is proposed that includes processing in the Fog via minified Kubernetes clusters located near the Edge, in the private Cloud via on-premises elastic clusters and, finally, in the public Cloud by using the AWS (Amazon Web Services) Lambda FaaS service. The results indicate that these strategies can mitigate overloads in use cases involving processing across the cloud-to-edge continuum by coordinating several layers of computing resources., Grant PID2020-113126RB-I00 funded by MCIN/AEI/10.13039/501100011033. Project PDC2021-120844-I00 funded by MCIN/AEI/10.13039/501100011033 and by the European Union NextGenerationEU/PRTR. This work was supported by the project AI-SPRINT "AI in Secure Privacy-Preserving Computing Continuum" that has received funding from the European Union's Horizon 2020 Research and Innovation Programme under Grant 101016577. This work was also supported by the project AI4EOSC "Artificial Intelligence for the European Open Science Cloud" that has received funding from the European Union's Horizon Europe Research and Innovation Programme under Grant 101058593.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

OSCAR-P and aMLLibrary: Profiling and predicting the performance of FaaS-based applications in computing continua

  • Sala, Roberto
  • Guindani, Bruno
  • Galimberti, Enrico
  • Filippini, Federica
  • Sedghani, Hamta
  • Ardagna, Danilo
  • Risco, Sebastián
  • Moltó, Germán|||0000-0002-8049-253X
  • Caballer Fernández, Miguel|||0000-0001-9393-3077
[EN] This paper proposes an automated framework for efficient application profiling and training of Machine Learning (ML) performance models, composed of two parts: OSCAR-P and aMLLibrary. OSCAR-P is an auto-profiling tool designed to automatically test serverless application workflows running on multiple hardware and node combinations in cloud and edge environments. OSCAR-P obtains relevant profiling information on the execution time of the individual application components. These data are later used by aMLLibrary to train ML-based performance models. This makes it possible to predict the performance of applications on unseen configurations. We test our framework on clusters with different architectures (x86 and arm64) and workloads, considering multi-component use-case applications. This extensive experimental campaign proves the efficiency of OSCAR-P and aMLLibrary, significantly reducing the time needed for the application profiling, data collection, and data processing. The preliminary results obtained on the ML performance models accuracy show a Mean Absolute Percentage Error lower than 30% in all the considered scenarios., The European Commission funded this work under the H2020 project AI-SPRINT (grant n. 101016577). OSCAR was developed under project PDC2021-120844-I00 funded by MCIN/AEI/-10.13039/5011000-11033 and by the European Union NextGenerationEU/PRTR and under grant PID2020-113126RB-I00 funded by MCIN/AEI/10.13039/501100011033. aMLLibrary was developed under the H2020 HPC JU LIGATE (grant n. 956137).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Leveraging an open source serverless framework for high energy physics computing

  • Padulano, Vincenzo Eduardo
  • Oliver Cortés, Pablo
  • Tejedor Saavedra, Enric
  • Risco, Sebastián
  • Alonso-Jordá, Pedro|||0000-0002-6882-6592
  • Moltó, Germán|||0000-0002-8049-253X
[EN] CERN (Centre Europeen pour la Recherce Nucleaire) is the largest research centre for high energy physics (HEP). It ofers unique computational challenges as a result of the large amount of data generated by the large hadron collider. CERN has developed and supports a software called ROOT, which is the de facto standard for HEP data analysis. This framework ofers a high-level and easy-to-use interface called RDataFrame, which allows managing and processing large data sets. In recent years, its functionality has been extended to take advantage of distributed computing capabilities. Thanks to its declarative programming model, the user-facing API can be decoupled from the actual execution backend. This decoupling allows physical analysis to scale automatically to thousands of computational cores over various types of distributed resources. In fact, the distributed RDataFrame module already supports the use of established general industry engines such as Apache Spark or Dask. Notwithstanding the foregoing, these current solutions will not be sufcient to meet future requirements in terms of the amount of data that the new projected accelerators will generate. It is of interest, for this reason, to investigate a diferent approach, the one ofered by serverless computing. Based on a frst prototype using AWS Lambda, this work presents the creation of a new backend for RDataFrame distributed over the OSCAR tool, an open source framework that supports serverless computing. The implementation introduces new ways, relative to the AWS Lambdabased prototype, to synchronize the work of functions., Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature. This work was supported by the research projects PID2020-113656RB-C22 (MCIN/AEI/10.13039/ 501100011033). Also, Grant PID2020-113126RB-I00 funded by MCIN/AEI/10.13039/501100011033 and Project PDC2021-120844-I00 funded by MCIN/AEI/10.13039/501100011033 funded by the European Union NextGenerationEU/PRTR all support the research on the OSCAR software tool.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Efficient and scalable covariate drift detection in machine learning systems with serverless computing

  • Céspedes Sisniega, Jaime
  • López García, Álvaro
  • Rodríguez-Benítez, Vicente|||0000-0002-7335-3849
  • Moltó, Germán|||0000-0002-8049-253X
[EN] As machine learning models are increasingly deployed in production, robust monitoring and detection of concept and covariate drift become critical. This paper addresses the gap in the widespread adoption of drift detection techniques by proposing a serverless-based approach for batch covariate drift detection in ML systems. Leveraging the open-source OSCAR framework and the open-source Frouros drift detection library, we develop a set of services that enable parallel execution of two key components: the ML inference pipeline and the batch covariate drift detection pipeline. To this end, our proposal takes advantage of the elasticity and efficiency of serverless computing for ML pipelines, including scalability, cost-effectiveness, and seamless integration with existing infrastructure. We evaluate this approach through an edge ML use case, showcasing its operation on a simulated batch covariate drift scenario. Our research highlights the importance of integrating drift detection as a fundamental requirement in developing robust and trustworthy AI systems and encourages the adoption of these techniques in ML deployment pipelines. In this way, organizations can proactively identify and mitigate the adverse effects of covariate drift while capitalizing on the benefits offered by serverless computing., This work was supported by the project AI4EOSC "Artificial Intelligence for the European Open Science Cloud" that has received funding from the European Union's Horizon Europe Research and Innovation Programme under Grant 101058593. JCS and ALG acknowledge the funding from the Agencia Estatal de Investigacion, Unidad de Excelencia Maria de Maeztu, ref. MDM-2017-0765. GM and VR would like to acknowledge Grant PID2020-113126RB-I00 funded by MICIU/AEI/10.13039/501100011033 and also project PDC2021-120844-I00 funded by MICIU/AEI/10.13039/501100011033 and by the European Union NextGenerationEU/PRTR.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Infrastructure Manager: A TOSCA-Based Orchestrator for the Computing Continuum

  • Caballer Fernández, Miguel|||0000-0001-9393-3077
  • Moltó, Germán|||0000-0002-8049-253X
  • Calatrava Arroyo, Amanda|||0000-0002-9018-9171
  • Blanquer Espert, Ignacio|||0000-0003-1692-8922
[EN] The edge-to-cloud continuum involves heterogeneous computing resources, including low-power physical devices, Virtual Machines (VMs) in cloud management platforms and serverless computing services based on the FaaS (Functions as a Service) model. This requires novel strategies to describe and efficiently deploy complex applications that execute across the computing continuum. To this end, this paper introduces the developments in the Infrastructure Manager (IM), an open-source TOSCA-based orchestrator to provision and configure virtualized computing resources from a wide range of cloud platforms. By supplementing TOSCA with additional types, the IM can also provision from FaaS platforms across the computing continuum by leveraging public cloud services such as AWS Lambda and on-premises serverless platforms, such as OSCAR. This allows event-driven data-processing applications across multiple computing platforms and architectures. The evolution of the Infrastructure Manager is described to accommodate the definition in TOSCA of complex applications that span across the computing continuum and their automated provisioning and configuration using Infrastructure as Code (IaC) approaches. Its effectiveness is assessed through a real use case involving a machine-learning classifier application for assisting in the early diagnosis of Rheumatic Heart Disease (RHD). The results show that the new developments enable the IM to efficiently deploy complete application architectures described in TOSCA across the computing continuum, from VMs to FaaS services., Grant PID2020-113126RB-I00 funded by MCIN/AEI/10.13039/501100011033. Also, Project PDC2021- 120844-I00 funded by MCIN/AEI/10.13039/501100011033 and by the European Union NextGenerationEU/PRTR. This work was partially supported by the project AI-SPRINT AI in Secure Privacy-Preserving Computing Continuum which has received funding from the European Union s Horizon 2020 Research and Innovation Programme under Grant 101016577 and by the AI4EOSC Artificial Intelligence for the European Open Science Cloud that has received funding from the European Union s Horizon Europe Research and Innovation Programme under Grant 101058593. Finally, Amanda wants to thank the support received by the Universitat Politècnica de València for the project Extensión de modelos de computaci¿n en cloud en entornos confiables y altamente distribuidos , from the PAID-10-20 postdoctoral program.
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

Efficient and scalable covariate drift detection in machine learning systems with serverless computing

  • Céspedes Sisniega, Jaime
  • Rodríguez Rodríguez, Vicente
  • Moltó, Germán
  • López-García, Álvaro
As machine learning models are increasingly deployed in production, robust monitoring and detection of concept and covariate drift become critical. This paper addresses the gap in the widespread adoption of drift detection techniques by proposing a serverless-based approach for batch covariate drift detection in ML systems. Leveraging the open-source OSCAR framework and the open-source Frouros drift detection library, we develop a set of services that enable parallel execution of two key components: the ML inference pipeline and the batch covariate drift detection pipeline. To this end, our proposal takes advantage of the elasticity and efficiency of serverless computing for ML pipelines, including scalability, cost-effectiveness, and seamless integration with existing infrastructure. We evaluate this approach through an edge ML use case, showcasing its operation on a simulated batch covariate drift scenario. Our research highlights the importance of integrating drift detection as a fundamental requirement in developing robust and trustworthy AI systems and encourages the adoption of these techniques in ML deployment pipelines. In this way, organizations can proactively identify and mitigate the adverse effects of covariate drift while capitalizing on the benefits offered by serverless computing., This work was supported by the project AI4EOSC “Artificial Intelligence for the European Open Science Cloud” that has received funding from the European Union’s Horizon Europe Research and Innovation Programme under Grant 101058593. JCS and ALG acknowledge the funding from the Agencia Estatal de Investigación, Unidad de Excelencia María de Maeztu, ref. MDM-2017-0765. GM and VR would like to acknowledge Grant PID2020-113126RB-I00 funded by MICIU/AEI/10.13039/501100011033 and also project PDC2021-120844-I00 funded by MICIU/AEI/10.13039/501100011033 and by the European Union NextGenerationEU/PRTR ., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

Rescheduling serverless workloads across the cloud-to-edge continuum

  • Risco, Sebastián
  • Alarcón, Caterina
  • Langarita, Sergio
  • Caballer, Miguel
  • Moltó, Germán
Serverless computing was a breakthrough in Cloud computing due to its high elasticity capabilities and fine-grained pay-per-use model offered by the main public Cloud providers. Meanwhile, open-source serverless platforms supporting the FaaS (Function as a Service) model allow users to take advantage of many of their benefits while operating on the on-premises platforms of organizations. This opens the possibility to deploy and exploit them on the different layers of the cloud-to-edge continuum, either on IoT (Internet of Things) devices located at the Edge (i.e. next to data acquisition devices), in on-premises clusters closer to the data sources (i.e. Fog computing) or directly on the Cloud. This paper presents two strategies to mitigate the overload that disparate data ingestion rates may cause in low-powered devices at the Edge or Fog layers. To this end, it is proposed to delegate and reschedule serverless jobs between the different layers of the cloud-to-edge continuum using an open-source platform for event-driven file processing. To demonstrate the performance of these strategies, a use case for fire detection is proposed that includes processing in the Fog via minified Kubernetes clusters located near the Edge, in the private Cloud via on-premises elastic clusters and, finally, in the public Cloud by using the AWS (Amazon Web Services) Lambda FaaS service. The results indicate that these strategies can mitigate overloads in use cases involving processing across the cloud-to-edge continuum by coordinating several layers of computing resources., Grant PID2020-113126RB-I00 funded by MCIN/AEI/10.13039/501100011033. Project PDC2021-120844-I00 funded by MCIN/AEI/10.13039/501100011033 and by the European Union NextGenerationEU/PRTR. This work was supported by the project AI-SPRINT “AI in Secure Privacy-Preserving Computing Continuum” that has received funding from the European Union’s Horizon 2020 Research and Innovation Programme under Grant 101016577. This work was also supported by the project AI4EOSC “Artificial Intelligence for the European Open Science Cloud” that has received funding from the European Union’s Horizon Europe Research and Innovation Programme under Grant 101058593., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

OSCAR-P and aMLLibrary: Profiling and Predicting the Performance of FaaS-based Applications in Computing Continua

  • Sala, Roberto
  • Guindani, Bruno
  • Galimberti, Enrico
  • Filippini, Federica
  • Sedghani, Hamta
  • Ardagna, Danilo
  • Risco, Sebastián
  • Moltó, Germán
  • Caballer, Miguel
This paper proposes an automated framework for efficient application profiling and training of Machine Learning (ML) performance models, composed of two parts: OSCAR-P and aMLLibrary. OSCAR-P is an auto-profiling tool designed to automatically test serverless application workflows running on multiple hardware and node combinations in cloud and edge environments. OSCAR-P obtains relevant profiling information on the execution time of the individual application components. These data are later used by aMLLibrary to train ML-based performance models. This makes it possible to predict the performance of applications on unseen configurations. We test our framework on clusters with different architectures (x86 and arm64) and workloads, considering multi-component use-case applications. This extensive experimental campaign proves the efficiency of OSCAR-P and aMLLibrary, significantly reducing the time needed for the application profiling, data collection, and data processing. The preliminary results obtained on the ML performance models accuracy show a Mean Absolute Percentage Error lower than 30% in all the considered scenarios.The European Commission funded this work under the H2020 project AI-SPRINT (grant n. 101016577). OSCAR was developed under project PDC2021-120844-I00 funded by MCIN/AEI/-10.13039/5011000-11033 and by the European Union NextGenerationEU/PRTR and under grant PID2020-113126RB-I00 funded by MCIN/AEI /10.13039/501100011033. aMLLibrary was developed under the H2020 HPC JU LIGATE (grant n. 956137)., The European Commission funded this work under the H2020 project AI-SPRINT (grant n. 101016577). OSCAR was developed under project PDC2021-120844-I00 funded by MCIN/AEI/-10.13039/5011000-11033 and by the European Union NextGenerationEU/PRTR and under grant PID2020-113126RB-I00 funded by MCIN/AEI /10.13039/501100011033. aMLLibrary was developed under the H2020 HPC JU LIGATE (grant n. 956137)., Peer reviewed