Edge computing devices have increased in number and capability over recent years. The ability to process data and execute machine learning in proximity to data generation and collection sources provides several advantages over using cloud- based data centers. We describe an orchestration mechanism that enables edge devices to make more effective use of energy resources in their proximity – a technique we refer to as “edge energy orchestration”. A software “orchestrator” can take account of renewable generation to alter how task execution on edge devices is carried out. An application scenario is used to illustrate the use of the orchestrator in practice, followed by a discussion about how this approach can be generalized to a broader set of applications

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

Edge Energy Orchestration

  • Vijay Kumar,
  • Nima Valizadeh,
  • Ioan Petri,
  • Omer Rana,
  • Ambarish Nag,
  • Sagi Zisman,
  • Charles Tripp,
  • Shuva Paul

摘要

Edge computing devices have increased in number and capability over recent years. The ability to process data and execute machine learning in proximity to data generation and collection sources provides several advantages over using cloud- based data centers. We describe an orchestration mechanism that enables edge devices to make more effective use of energy resources in their proximity – a technique we refer to as “edge energy orchestration”. A software “orchestrator” can take account of renewable generation to alter how task execution on edge devices is carried out. An application scenario is used to illustrate the use of the orchestrator in practice, followed by a discussion about how this approach can be generalized to a broader set of applications