What is it about?
Workload forecasting is important for edge computing. We have shown for the first time that accurate and time-efficient workload forecasting can be achieved in a collaborative cloud-edge manner. Extensive evaluation utilizing the realistic workload datasets collected from Alibaba ENS further demonstrates the effectiveness of our methods.
Featured Image
Photo by NASA on Unsplash
Why is it important?
The first study leverages the collaborative cloud-edge paradigm for edge workload forecasting. We propose ELASTIC, which not only captures the intra-site correlations but also the inter-site correlations. Extensive evaluation utilizing the realistic workload datasets collected from Alibaba ENS demonstrates the effectiveness of ELASTIC.
Perspectives
Writing this article was a great pleasure as it has co-authors with whom I have had long-standing collaborations. We want to be grounded in real edge computing scenarios and strive to make a continuous contribution.
Yanan Li
Read the Original
This page is a summary of: ELASTIC: Edge Workload Forecasting based on Collaborative Cloud-Edge Deep Learning, April 2023, ACM (Association for Computing Machinery),
DOI: 10.1145/3543507.3583436.
You can read the full text:
Resources
Contributors
The following have contributed to this page
