Project Description

Phoeβe (/ˈfiːbi/) wants to add basic artificial intelligence capabilities to the Linux OS.

System-level tuning is a very complex activity, requiring the knowledge and expertise of several (all?) layers which compose the system itself, how they interact with each other and (quite often) it is required to also have an intimate knowledge of the implementation of the various layers.

Another big aspect of running systems is dealing with failure. Do not think of failure as a machine turning on fire rather as an overloaded system, caused by misconfiguration, which could lead to starvation of the available resources.

In many circumstances, operators are used to deal with telemetry, live charts, alerts, etc. which could help them identifying the offending machine(s) and (re)act to fix any potential issues.

However, one question comes to mind: wouldn't it be awesome if the machine could auto-tune itself and provide a self-healing capability to the user? Well, if that is enough to trigger your interest then this is what Phoeβe aims to provide.

Phoeβe uses system telemetry as the input to its brain and produces a big set of settings which get applied to the running system. The decision made by the brain is continuously reevaluated (considering the grace_period setting) to offer eventually the best possible setup.

Goal for this Hackweek

Work mostly on two main areas:

1) Rework the data engineering part of Phoebe to add tags/labels to individual data field to be used by the model;

2) Update the model according to the data re-engineering

3) Create a tool to assist Phoebe with data manipulation so to move away from CSV files

Stretch goal: have a proper lab setup to consistently test and validate Phoebe and generate data.

Resources

URL: https://github.com/SUSE/phoebe

Events in calendar

Monday 22nd March 2021 @ 10:00 AM CEST - Meeting with Prof. Nicola Strisciuglio

Every day @ 9:00 AM CEST - Sync up on progress, opens and... have a coffee together :)

Join this project Leave this project

Looking for hackers with the skills:

linux artificial-intelligence machinelearning c python meson ci/cd tuning self-healing performance reliability

This project is part of:

Hack Week 20

Activity

over 2 years ago: asmorodskyi joined this project.

almost 3 years ago: llansky3 liked this project.

almost 3 years ago: dfaggioli liked this project.

almost 3 years ago: hennevogel liked this project.

almost 3 years ago: ybonatakis joined this project.

almost 3 years ago: ybonatakis liked this project.

almost 3 years ago: mlnoga liked this project.

almost 3 years ago: tjyrinki_suse liked this project.

almost 3 years ago: mvarlese added keyword "reliability" to this project.

almost 3 years ago: mvarlese added keyword "performance" to this project.

almost 3 years ago: mvarlese added keyword "self-healing" to this project.

almost 3 years ago: mvarlese added keyword "tuning" to this project.

almost 3 years ago: dmulder joined this project.

almost 3 years ago: dancermak liked this project.

almost 3 years ago: shunghsiyu joined this project.

almost 3 years ago: shunghsiyu liked this project.

almost 3 years ago: mvarlese joined this project.

almost 3 years ago: mslacken started this project.

almost 3 years ago: mvarlese added keyword "c" to this project.

almost 3 years ago: mvarlese added keyword "python" to this project.

almost 3 years ago: mvarlese added keyword "meson" to this project.

almost 3 years ago: mvarlese added keyword "ci/cd" to this project.

almost 3 years ago: mvarlese added keyword "machinelearning" to this project.

almost 3 years ago: mvarlese added keyword "linux" to this project.

almost 3 years ago: mvarlese added keyword "artificial-intelligence" to this project.

Comments

almost 3 years ago by dmulder | Reply

I'd like to see if I could use Phoebe to tune samba settings (long term goal). So, this hackweek I think I'll just familiarize myself with Phoebe.

almost 3 years ago by mvarlese | Reply

That's awesome!!! A can see a new plugin coming to Phoebe :)

almost 3 years ago by mlnoga | Reply

Great one, Marco. Two points to consider:

What are the top 4-5 settings to tune?
What kind of data is available for the team in HackWeek to learn settings? (sourcing during Hackweek is probably not a viable option)

almost 3 years ago by mvarlese | Reply

Thanks for the feedback Markus! Will keep both points in mind.

Project Description

Lesser on the coding s...

Authenticated hashes for BTRFS by dsterba