PhD position: Distributed computing applied in a satellite swarm system

*Keywords*: orbital edge computing, distributed algorithms, satellite swarm, nanosatellites, replication, distributed computing, fault tolerance, intermittent computing, inter-satellite network.

*Context*

It is fairly common today for on-the-ground applications to make use of high performance computing (HPC) infrastructures, either on premises or using cloud services, that allow for the computation of highly-demanding processes. One of the key technology behind these infrastructures is the use of massively parallel and distributed computing.

Considering the deployment of more and more powerful processing units onboard satellites, and in particular of mainstream commercial-off-the-shelves (COTS), the capability for more heavy computing directly in space has become a growing trend. A new paradigm in space system, namely swarm of satellites, might open a new way to answer this trend and fill the gap between what is offered with distributed computing with HPC on Earth and what is possible in space systems.

Satellite swarms are space systems constituted of multiple satellites (usually nanosatellites), from 2-3 to tens or hundreds of satellites, cooperating with each other as one entity to accomplish a specific mission (earth observation, science, defence, etc.). To accomplish such application, each satellite is equipped with Inter-Satellite link (ISL) communication means (either through optical or radio-frequency links) allowing communication between each node of the swarm through an appropriate network infrastructure. Such swarm missions concepts have begun to raise with preliminary scientific projects such as NOIRE[CDB + 18], SMOS-Next, ULID, DROID and the need for proposing a new kind of processing infrastructure becomes an obvious challenge.

Considering more and more demanding on-board applications such as autonomous orbit control with collision risk and formation management, image processing, AI, etc, it is natural to consider exploiting the power of every processing unit hosted in each satellite of the swarm as we would distribute a process between nodes of an HPC on Earth. However, constraints in space swarms or constellations are very different from its on-the-ground counterpart as described in the next chapter.

  * Survey the technologies available on the ground or in space application (e.g., [DKL + 17, DL20]) for distributed computing and consider them in the face of swarm system constraints

  * Identify, depending on the different industrial or scientific use cases and scenarios brought by CNES or other partners, how to apply in swarm systems such technologies for distributed computing and/or investigate how to adapt them to this kind of environment and applications.

  * Confront the hypotheses through demonstration on the CNES swarm experimental facilities.

*Proposed research*

The PHD thesis will aim at tackling the following scientific challenges:

  * How to apply distributed computing to satellite swarm systems considering the specific constraints that do not apply for Earth counterparts ? This include a very limited data transmission speed (a few kb/s to 1Mb/s for instance), intermittent connexion between nodes, possible loss or corruption of data, limited CPU power (700MHz dual core Arm processor for instance), … Identify or design the technologies (framework, software components, protocols, etc) to perform such computing.

  *At system level, how the different technologies for distributed computing in a satellite swarm may or may not apply to the different industrial use cases and their needs (high resiliency or not, reactivity, CPU load, etc.) ? Evaluate performances, limitations, advantages and drawbacks compared to standard non-distributed computing. This work will rely on demonstrations on a representative swarm testbed with nanosatellite-compliant hardware provided by CNES.

  * How to operate, at system level, such technologies? Evaluate the degrees of autonomy and resiliency of such processes, how to monitor it and mitigation techniques to apply.

Currently, the availability and fault-tolerance of traditional, cloud-based distributed system are commonly guaranteed by a replication protocol based on replicated state machine (RSM). Such a protocol implements a consensus algorithm to enable strong consistency, like Fast Paxos [Lam06] and Raft [OO14]. Strongly consistent replication is key to efficient implementation of critical distributed systems’ building blocks, like distributed lock manager, reliable configuration or transactional key-value store. To our knowledge though, such protocols have never been designed and extensively evaluated on nanosatellite constellations. Finally, it is important to highlight that this doctoral project is a joint work with our colleagues at the National Centre for Space Studies (CNES).

*Requirements and application*

In this research project, we intend to explore both a fundamental and an applied aspects. Candidates to this position should hold a Master’s degree in Computer Science/Informatics, Mathematics or a related field by the starting date of the doctoral project. They must be excited by research in distributed systems/computing, distributed algorithms, orbital edge computing, and/or intermittent computing, and should have an excellent academic record in one of these areas. Familiarity with formal specification and verification, and graph theory/algorithms would be greatly appreciated. Teamwork and communication skills are key to this position, and industrial experience is a plus.

Excellent proficiency in English is required (CECR : C1; IELTS : 7.0; Cambridge English Scale : 185; or equivalent). Knowledge of French is not required for this position.

To apply, please send the following information to [email protected](Subject=PhD position [ENAC-CNES-PhD24]: distributed computing in nanosats swarms):

  * Curriculum Vitæ

  * Letter of motivation that should describe the applicant's background in the areas of the project, reason for interest in the project, and future plans

  * A list of courses and grades of the last three years of study (an informal transcript is OK).

  * Names and contact details of at least two people who can write you references, whom we will contact directly.

  * If relevant, a link to your publications and/or open-source developments.

Application deadline: 20 February 2024.

This fully-funded PhD starts in October 2024 and the duration of the contract/scholarship is 3 years. This project is fully funded by CNES.

*Eligibility criteria and Benefits*

Since CNES conducts strategic activities in the space domain for the French government, applicants' eligibility is subject to the result of a PPST procedure. Still, applicants of any nationality can apply, but applicants must not have a doctoral degree already or been enrolled in a PhD/doctoral program.

Benefits include:

  * French government strongly subsidizes its higher education system, therefore our program grants tuition fee waiver at Université de Toulouse

  *  Doctoral students are eligible for an accommodation in our own campus which contributes significantly to reduce the total cost of living (already much cheaper than bigger European cities)

  * Social security coverage included

  * Subsidized meals 

  * Partial reimbursement of public transport costs

  * Social, cultural and sports events and activities

*About CNES and ENAC*

Founded in 1961, National Centre for Space Studies (CNES) is the government agency responsible for shaping and implementing France’s space policy in Europe. To execute the nation’s space policy, CNES relies on strong shared values like excellence, enthusiasm and the desire to rise to the challenges that lie ahead. Our 2,357 men and women are working hard to lay plans for the future, make French industry more competitive and sustain scientific and operational excellence.

The ENAC, National School of Civil Aviation, is located in Toulouse, France, the centre of the European aerospace industry (e.g., AirBus, Thales, and CNES). It offers an ideal working environment, where researchers can focus on developing new ideas, collaborations and projects. The research topics at ENAC Lab include emerging CPS design (e.g., drones and nanosatellites), aviation safety and security, sustainable transportation development, and aeronautical computer-human interactions. For further information, please consult our site.

The proposed research will be developed in the CNES site in Toulouse, France, in close cooperation with ENAC research laboratory, ENAC Lab.

*References*

[CDB + 18] B. Cecconi et al. "NOIRE study report: towards a low frequency radio interferometer in space." 2018 IEEE Aerospace Conference. IEEE, 2018.

[DKL + 17] K. Devaraj et al. Dove high speed downlink system. 2017.

[DL20] B. Denby and B. Lucia. Orbital edge computing: Nanosatellite constellations as a new class of computer system. In the ASPLOS, 2020.

[Lam06] L. Lamport. Fast paxos. Distributed Computing, 2006.

[OO14] D. Ongaro and . Ousterhout. In search of an understandable consensus algorithm. In the ATC, 2014.

Website: http://perso.recherche.enac.fr/~silvestre/jobs.html#nanosats1