Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

Sl. No.TopicPresenterNotes
1

EUAG Paper: Action Item Discussions

  •  Platforms
  •  Related Projects
  •  Data Modeling
  •  Repository of Problems
  •  Test and Certify.
  •  Training

Platform:  Data Pipeline? ML-Models ? Language/libraries ? Integration-Flexibility ? 

Project: Explore O-RAN.

Data Modeling:  (a) Sources (b) time-series (c) logs (d) understanding of the columns (e) Terminologies used  (f) naming consistency. Can YANG help ? (To Explore). Does YANG already cover metrics/logs - infrastructure monitoring ? Is there any platform which is normalizing multiple-sources and having a common data model?

Repository of the problems (use-cases).

Example of where bias in data can have unintended consequences: https://courses.cs.duke.edu//spring20/compsci342/netid/readings/facialrecnytimes.pdf

Test/Certify: ML-Model. Against Common Training + Test Data.  Metrics: Speed + Accuracy. Publish: Training and expected metrics. Richness of the data-set is very important.

Training: Amount of training available on AI/ML and Networking in public is HUGE. LF-Course?\

https://www.prnewswire.com/news-releases/att-and-h2oai-launch-co-developed-artificial-intelligence-feature-store-with-industry-first-capabilities-301410998.html 

2Kubernetes Failure Emulation

Hypothesis: 

We cannot emulate failure of a pod, by running stress tools. CPU loads have no impact. Even if you allocate 2GB RAM (with static configuration), and do memory operation (stressng) with buffer size of 4 to 6 GB, still there will be no failures. Tried these configurations: https://github.com/opensource-tnbt/stressng-images.

What is more important ML-problem for K8S for NFV usecases is an open question ?

3GANs for Synthetic Data Generation

Moved to Next week due to non-availability of the students.
4BERT for Openstack Log Analysis
5

Webinar, Testing Forum




Role of the Projects such as Thoth.

  • Intelligent Networking webinar - ~45 minutes prepared material and ~15 of Q&A
  • Focus on what has been done in support of the findings
  • Propose inviting Sridhar Rao from Anuket's Thoth project to participate in the webinar
  • Panelists: Beth Cohen,  Massimo Banzi,  Sridhar Rao Lei Huang Yuhan Zhang  
  • Recommendation for early December date