Many science and industry IoT applications necessitate data processing across the edge-to-cloud continuum to meet performance, security, cost, and privacy requirements. However, diverse abstractions and infrastructures for managing resources and tasks across the edge-to-cloud scenario are required. We propose Pilot-Edge as a common abstraction for resource management across the edge-to-cloud continuum. Pilot-Edge is based on the pilot abstraction, which decouples resource and workload management, and provides a Function-as-a-Service (FaaS) interface for application-level tasks. The abstraction allows applications to encapsulate common functions in high-level tasks that can then be configured and deployed across the continuum. We characterize Pilot-Edge on geographically distributed infrastructures using machine learning workloads (e.g., k-means and auto-encoders). Our experiments demonstrate how Pilot-Edge manages distributed resources and allows applications to evaluate task placement based on multiple factors (e.g., model complexities, throughput, and latency).
翻译:许多科学和工业IoT应用程序需要跨边对边连续运行的数据处理,以满足业绩、安全、成本和隐私要求。然而,需要不同的抽象和基础设施来管理边缘对边假设情况下的资源和任务。我们提议试点切换,作为跨边对边连续运行资源管理的共同抽象。试点切换以试点抽取为基础,它分解资源和工作量管理,并为应用层面的任务提供一个功能即服务界面。抽象化使得应用程序能够将共同功能包含在高层次的任务中,然后可以配置并在整个连续运行中部署。我们用机器学习工作量(例如K-手段和自动集成器)对地理分布的基础设施进行定性。我们的实验表明试点切换如何管理分配的资源,并允许应用程序根据多种因素(例如模型复杂性、吞吐量和静态)评估任务安排。