Fault and Performance Management in Multi-Cloud Based NFV using Shallow and Deep Predictive Structures
A presentation related to this paper
at the 7th Workshop on Industrial Internet of Things Communication Networks (IioTCom) at ICCCN 2017, Vancouver, Canada, August 3, 2017
This talk covers the following topics:
- Network Function Virtualization
- Advantages of NFV
- Trend: Computation in the Edge
- Trend: Multi-Cloud
- Advantages of NFV on Multi-Cloud
- Issues in Multi-Cloud NFV Deployments
- Network Services (NS)
- FCP Problem Description
- Markers and Metrics
- Description of Training Datasets
- Telstra Dataset Samples
- KDE dataset
- Fault Detection
- Detection of Faults and Performance Problems
- Detection (cont.)
- Localization of Faults
- Localization of Faults and Performance Problems
- Stacked Autoencoder
- Confusion Matrix
- Effect of Relative Sizes of Hidden Layers
- References
- Scan This to Download These Slides
Presentation slides in Adobe Acrobat Format:
1 slide/page
| 4 slides/page
View
Slides+Audio
Right click and save to download
Slides+Audio podcast.
Frequently asked questions about remote audio/video viewing
Back to other lectures of the series
Complete List of Audio/Video Lectures by Raj Jain
Back to Raj Jain's Home Page