PhD Position F/M PhD Position: Understanding Linux scheduling bottlenecks

Inria

France

November 15, 2022

Description

2022-05378 - PhD Position F/M PhD Position: Understanding Linux scheduling bottlenecks

Contract type : Fixed-term contract

Level of qualifications required : Graduate degree or equivalent

Fonction : PhD Position

Level of experience : Recently graduated

About the research centre or Inria department

The Inria Rennes - Bretagne Atlantique Centre is one of Inria's eight centres and has more than thirty research teams. The Inria Center is a major and recognized player in the field of digital sciences. It is at the heart of a rich R&D and innovation ecosystem: highly innovative PMEs, large industrial groups, competitiveness clusters, research and higher education players, laboratories of excellence, technological research institute, etc.

Context

The Linux kernel is at the core of almost all modern computing, from the smartphone to the datacenter. As such, its performance and correctness are critical. In a virtualized environment, such as the cloud, operating system functionalities are taken over by the hypervisor, which can be viewed as the new operating system. Xen and Linux-KVM are the two most popular open source hypervisors used by large scale cloud providers such as Amazon and Outscale. In order to ensure e ective and e cient support for modern applications, it is necessary to understand and address bottlenecks in both traditional operating systems and hypervisors, as well as in the interplay between them. This Phd position is part of a research project in operating systems being carried out in the {Whisper} and {WIDE} Inria teams.

The PhD Position will take place at Inria Rennes with a tight collaboration with the {Whisper} Inria team in Paris.

Assignment

Context:

The process scheduler is the part of an operating system that decides what thread runs on what core at what time. As such, it has a critical impact on application performance, particularly for multithreaded applications. While some applications simply spread out across the available cores, with each thread running continuously without disturbance on its preferred core, others involve frequent synchronization, I/O, etc. In practice, whenever a thread gives up access to a core, another thread can replace it, leading to migrations, loss of locality, and degraded performance. At the same time, evolutions in the scheduler can introduce errors, that can also degrade performance on specific workloads. Tools exist to trace scheduling behavior, that can help identify the presence of such problems, but the sheer volume of information available makes it difficult to map a scheduling trace to a root cause.

Objective:

Heterogeneous ISAs (Instruction Set Architectures) are greatly gaining in popularity, as illustrated by

the new generation of Intel processors and Apple's A and M processors that bring the big.LITTLE CPU principle from ARM to desktop use. Although the idea of leveraging a heterogeneous ISA is not entirely novel,

it brings new challenges for schedulers as illustrated by the disappointing performance observed with Intel's patches to the Linux scheduler to support their latest Alder Lake architecture. With Alder Lake, Intel introduces P-core and E-core, each with different levels of performance and targeted at different sorts of workloads. In particular,

P-cores are SMT capable, whereas E-cores are not. Recent changes to the Linux scheduler's management of thread affinity, to improve cache usage, turns out to cause active threads to become trapped on the less performant E-cores.

The goal of this PhD is to detect such scheduling problems quickly, to understand performance degradations and determine under which conditions performance is lower and for what reasons. Based on the detected problems, the PhD candidate will consider how to improve the Linux scheduler to provide better performance. Finally, one objective is also to analyze to what extent we can exploit the heterogeneity of an architecture to propose a system interface for applications to optimize performance while ensuring a reduced energy consumption. Designing such an interface poses challenges at several levels (hardware, system, and software engineering).

Expected background:

This PhD requires familiarity with the C language and operating systems concepts, as well as a background in machine learning.

Collaborations:

The research work will be carried out with Julia Lawall (https: // who.paris.inria.fr/Julia.Lawall/), Inria Senior Research Scientist (Directrice de Recherche) in the Whisper group at Inria-Paris

Skills

Technical skills and level required : Strong technical skills in the C language and Operating systems concepts

Languages : English

Other valued appreciated : A background in machine learning would be a plus.

Benefits package
  • Subsidized meals
  • Partial reimbursement of public transport costs
  • Possibility of teleworking ( 90 days per year) and flexible organization of working hours
  • partial payment of insurance costs
  • Remuneration

    monthly gross salary amounting to 2051 euros for the first and second years and 2158 euros for the third year

    General Information
  • Theme/Domain : Distributed Systems and middleware System & Networks (BAP E)

  • Town/city : Rennes

  • Inria Center : CRI Rennes - Bretagne Atlantique
  • Starting date : 2022-10-01
  • Duration of contract : 3 years, 1 month
  • Deadline to apply : 2022-11-15
  • Contacts
  • Inria Team : WIDE
  • PhD Supervisor : Bromberg David / david.bromberg@irisa.fr
  • The keys to success

    This PhD requires familiarity with the C language and operating systems concepts, a background in machine learning would be a plus.

    About Inria

    Inria is the French national research institute dedicated to digital science and technology. It employs 2,600 people. Its 200 agile project teams, generally run jointly with academic partners, include more than 3,500 scientists and engineers working to meet the challenges of digital technology, often at the interface with other disciplines. The Institute also employs numerous talents in over forty different professions. 900 research support staff contribute to the preparation and development of scientific and entrepreneurial projects that have a worldwide impact.

    Instruction to apply

    Please submit online : your resume, cover letter and letters of recommendation eventually

    For more information, please contact david.bromberg@inria.fr ou francois.taiani@inria.fr

    Defence Security : This position is likely to be situated in a restricted area (ZRR), as defined in Decree No. 2011-1425 relating to the protection of national scientific and technical potential (PPST).Authorisation to enter an area is granted by the director of the unit, following a favourable Ministerial decision, as defined in the decree of 3 July 2012 relating to the PPST. An unfavourable Ministerial decision in respect of a position situated in a ZRR would result in the cancellation of the appointment.

    Recruitment Policy : As part of its diversity policy, all Inria positions are accessible to people with disabilities.

    Warning : you must enter your e-mail address in order to save your application to Inria. Applications must be submitted online on the Inria website. Processing of applications sent from other channels is not guaranteed.

    Similar Jobs

    Inria

    France Sep 17, 2022

    Add to favorites Read more...

    PhD Position F/M Fairness in Federated Learning

    2022-05297 - PhD Position F/M Fairness in Federated Learning About the research centre or Inria department The Inria Lille - Nord Europe Research Centre was founded in 2008 and employs a staff of 360, including...