Scholarlink

<p>Serverless computing is a paradigm in which the underlying infrastructure is fully managed by the provider, enabling applications and services to be executed with elastic resource provisioning and minimal operational overhead. A core model within this paradigm is Function-as-a-Service (FaaS), where lightweight functions are deployed and triggered on demand, scaling seamlessly with workload. FaaS offers flexibility, cost-effectiveness, and fine-grained scalability, qualities particularly relevant for large-scale scientific infrastructures where data volumes are too large to centralise and computation must increasingly occur close to the data. The Square Kilometre Array Observatory (SKAO) exemplifies this challenge. Once operational, it will generate about 700&#xa0;PB of data products annually, distributed across the SKA Regional Centre Network (SRCNet), a federation of international centres providing storage, computing, and analysis services. In such a context, FaaS offers a natural mechanism to bring computation to the data. We studied the principles of Function-as-a-Service (FaaS) and their application to radio astronomy workflows. To demonstrate feasibility, a <i>Gaussian convolution</i> function was fully implemented and integrated within the SRCNet ecosystem. The methodology included a basic evaluation through performance tests to measure execution latency, comparing traditional computing environments against the FaaS-based approach deployed within the SRCNet distributed data and computing infrastructure. Benchmarks demonstrate that the FaaS framework, supported by an SRCNet IVOA Datalink service, successfully enables transparent function execution at the node closest to the data and user. These results show a measurable reduction in execution latency and inter-node data transfers, significantly enhancing overall computational efficiency within the SRCNet architecture. The SRCNet FaaS model, although currently under development, provides a viable pathway for seamless function deployment across a federated ecosystem. By leveraging SRCNet IVOA Datalink for high-performance, data-proximate computation, the system aligns with the SRCNet vision. However, transitioning to larger-scale workloads will necessitate precise resource sizing and high-availability scaling of the underlying functional nodes.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Bringing Computation to the Data: Interoperable Serverless Function Execution for Astrophysical Data Analysis in the SRCNet

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月4日

查看全文
  • Manuel Parra-Royón,
  • Julián Garrido-Sánchez,
  • Susana Sánchez-Expósito,
  • María Ángeles Mendoza,
  • Rob Barnsley,
  • Anthony Moraghan,
  • Jesús Sánchez,
  • Laura Darriba,
  • Carlos Ruíz-Monje,
  • Edgar Joao,
  • Javier Moldon,
  • Jesús Salgado,
  • Lourdes Verdes-Montenegro

摘要

Serverless computing is a paradigm in which the underlying infrastructure is fully managed by the provider, enabling applications and services to be executed with elastic resource provisioning and minimal operational overhead. A core model within this paradigm is Function-as-a-Service (FaaS), where lightweight functions are deployed and triggered on demand, scaling seamlessly with workload. FaaS offers flexibility, cost-effectiveness, and fine-grained scalability, qualities particularly relevant for large-scale scientific infrastructures where data volumes are too large to centralise and computation must increasingly occur close to the data. The Square Kilometre Array Observatory (SKAO) exemplifies this challenge. Once operational, it will generate about 700 PB of data products annually, distributed across the SKA Regional Centre Network (SRCNet), a federation of international centres providing storage, computing, and analysis services. In such a context, FaaS offers a natural mechanism to bring computation to the data. We studied the principles of Function-as-a-Service (FaaS) and their application to radio astronomy workflows. To demonstrate feasibility, a Gaussian convolution function was fully implemented and integrated within the SRCNet ecosystem. The methodology included a basic evaluation through performance tests to measure execution latency, comparing traditional computing environments against the FaaS-based approach deployed within the SRCNet distributed data and computing infrastructure. Benchmarks demonstrate that the FaaS framework, supported by an SRCNet IVOA Datalink service, successfully enables transparent function execution at the node closest to the data and user. These results show a measurable reduction in execution latency and inter-node data transfers, significantly enhancing overall computational efficiency within the SRCNet architecture. The SRCNet FaaS model, although currently under development, provides a viable pathway for seamless function deployment across a federated ecosystem. By leveraging SRCNet IVOA Datalink for high-performance, data-proximate computation, the system aligns with the SRCNet vision. However, transitioning to larger-scale workloads will necessitate precise resource sizing and high-availability scaling of the underlying functional nodes.