What is it about?
It studies the power efficiency and cost effects of controlling the utilization of multiple edge data centers that are interconnected between each other and towards a centralized cloud. The study incorporates server efficiency aspects as well as energy consumption coming from cooling but also spans energy being consumed by sending data across the network
Featured Image
Why is it important?
Edge computing is seen as a very important future evolution, but concerns are raised on the overall systemic efficiency as edge data centers risk being under-utilized as they are smaller and therefore will not experience statistical multiplexing gains from serving larger areas. So the ability and policies by which service providers can control the utilization and thereby the efficiency are crucial to achieving the needed economy of scale.
Perspectives
The aspect of moving data over large distances to central but quite efficient computing can be questioned when the amount of data grows exponentially. So instead moving computation to the source or destination of data is a good idea. But such approaches need careful optimization to achieve the needed efficiency
Mats Eriksson
Read the Original
This page is a summary of: Cost Optimization for the Edge-Cloud Continuum by Energy-Aware Workload Placement, June 2023, ACM (Association for Computing Machinery),
DOI: 10.1145/3599733.3600253.
You can read the full text:
Contributors
The following have contributed to this page
