In the context of global warming, reducing the carbon emissions of computer infrastructures is a top priority. Datacenters are responsible for a large fraction of the carbon emissions of IT.
Datacenters are at the core of Cloud infrastructures where most online services that we use in our every day life are hosted. Users of these services expect high quality of service [1]. This includes high availability where services should always be ready to answer user requests no matter the failures experienced by the Cloud infrastructure [2]. However, providing such guarantees has a high cost in terms of carbon footprint. Indeed, it mostly rely on different replication techniques that increase the hardware resources required to run services.
As of today, very few work have studied the link between the quality of service provided by such platforms and their carbon footprint [3]. Actually, we think that the situation on the planet requires us to look at the problem from a different angle: given a maximum carbon budget that we can afford, what quality of service can be ensured for an online service?
The goal of this internship is to study the carbon footprint of the fault tolerance strategies used by cloud providers today, to try identifying the strategies that are the best from environmental point of view.
The intern will be in charge of the following tasks:
Studying recent publications and collecting information about the strategies used by cloud provider to increase their quality of service.
Propose an approach (or several directions) to fault tolerance in the cloud under limited carbon budget
If possible, deploy different fault tolerance software solution on our private cloud infrastructure to validate experimentally the proposed approach
[1] Mogul and Wilkes. “Nines are not enough: Meaningful metrics for clouds.” (2019).
[2] Garraghan, Yang, et al. “Emergent failures: Rethinking cloud reliability at scale” (2018).
[3] Qiu, Haoran, et al. “When green computing meets performance and resilience SLOs.” (2024).
The intern will join the Informatics Laboratory of Univ. Grenoble Alpes (LIG), one of the largest laboratory in Computer Science in France. Univ. Grenoble Alpes is one of the 50 best universities in the world for computer science according to international rankings.
More specifically, the intern will work in the KrakOS research team, that focus on the efficiency and robustness of systems. KrakOS is one of the top system research teams in France and in Europe, with publications in the most selective conferences of the domain.