Points to Consider:
- This makes a lot of tasks within Observability compliant with licenses
- Certain tasks can be customized to create Custom Metrics
- Working with eBPF kernels will improve Encapsulation of Network
Packets, Data and Functions
- Consider AI tasks that require personal data processing, eBPF comes
into place to assist in Observability
- Usage of logging, tracing and metrics are recommended by EU GDPR
rules
- In logging , EU GDPR rule states: an identifier can be in the logs but not
an image such as a face image
- A Software Development Lifecycle (SDLC) has much larger span than
a Hardware Development Lifecycle (HDLC)
- A Data Lifecycle (DLC) is even larger than SDLC
- Consider we do Metrics of Face Recognition on a Linux Kernel (eBPF)
- By Face Age : The prediction of a Face using an ML model, gives 94% accuracy with the Real Drift Detection technique using a Classifier

Availability:
- eBPF on Android with Linux Kernel
- eBPF on Cameras with Linux Kernel
- eBPF on Servers with Linux Kernel

Standardization of Data Delivery:
- More Privacy in Images, such as Personal Data involved in text format,
a Passport for example, requires Image Classification Models with
Policy Integrated
- Lesser Privacy Images, require no Policy at all within the Models

свернуть