KUBILAI - Kubernetes Simplified with AI-powered Automatic Scaling

Start: 01/01/2024
End: 31/05/2026
Funding: Catalan
Status: On going
Research unit:
Services as Networks (SaS)
Acronym: KUBILAI

According to a study published by Lawrence Berkeley National Laboratory in 2016, data centers and their associated networks account for approximately 1% of global electricity consumption and generated around 160 million metric tons of carbon dioxide (CO2) emissions per year which is 0.3% of global CO2 emissions. This is equivalent to (i) The entire aviation industry in terms of CO2 emissions or (ii) The annual emissions of about 34 million cars or (iii) The total CO2 emissions of the entire country of Spain. The study estimated that data centers and their networks could consume up to 13% of global electricity by 2030 if trends continue, resulting in even higher CO2 emissions.

To address these challenges, we propose a game-changing software solution that employs AI-based approaches to establish highly-automated optimal configurations for both scaling and tuning of cloud applications. By using KUBILAI software, enterprises can reduce the carbon emissions produced by redundant servers and minimize unnecessary resource consumption, while still ensuring optimal performance and user experience. Moreover, KUBILAI’s aim is to democratise this solution and make it accessible to any company, in order to contribute to solving carbon emission problem of our planet. KUBILAI software leverages machine learning algorithms to continuously monitor the application and its environment, such as resource usage, user traffic, and workload patterns. It then uses this data to automatically determine the optimal resource allocation, scaling, and tuning strategies to ensure optimal performance, stability, and SLA adherence. KUBILAI software is fully integrated to Kubernetes, therefore it will work across various cloud platforms and environments.

In addition to providing a highly-automated approach to cloud application management, KUBILAI solution optimizes cloud resources to minimize costs while still ensuring optimal performance and user experience. It promotes environmental sustainability by reducing unnecessary resource consumption and carbon emissions. By reducing overcommitment of resources, KUBILAI solution can significantly reduce carbon emissions and contribute to a more sustainable future.

Engin Zeydan
PI/Project Leader
Akram Galal
Miquel Payaró
Josep Mangues
No results found
No results found