Doctorate in Applied and Engineering Physics  

Home
School
login
About the Program
Program Structure
Research Centers
Admission
Application
Scholarships
Contacts

Thesis Proposal for the

Doctorate Program in Applied and Engineering Physics (DAEPHYS)

Starting in the Academic Year 2016/2017


Proposal 027


Title:Development of accelerated jet substructure parallel algorithms using graphical processing units for the ATLAS upgrade
 
Objectives (recommended length: 2000-3000 char):
In this PhD project, the student will use the power of graphical processing units (GPUs) to enhance the processing speed of jet substructure algorithms employed in the ATLAS experiment at the LHC. If successful, this will be a very important technological development, which will allow to efficiently analyse the vast amounts of data expected of the high-luminosity LHC phase (HL-LHC) within the limitations imposed by expected computing resources.
 
Framework (recommended length: 500-2000 char):
Hadronic jets are the fingerprints of energetic quarks and gluons produced in particle physics experiments. They are crucial instruments in a wide range of subjects, from precision measurements of the inner structure of the proton, to Higgs boson measurements, or the search for heavy particles predicted by new physical theories. The field of jet reconstruction has undergone an enormous boost in the last few years, especially through the development of jet substructure techniques, capable of identifying heavy resonance decays into a single jet. With the enhanced LHC centre of mass energy in the current run, these techniques are now at the forefront of the experimental searches.

The very high luminosity expected of future LHC runs, and especially of the HL-LHC phase, will make it increasingly hard to deploy enough computing resources for the experimental needs. Indeed, the estimated increases in trigger rates and event size lead to much longer event reconstruction times, that are not matched by the slower expected growth in computing power at fixed cost. This implies that a change in paradigm is needed. How this will happen is at present unknown, but it is clear that the solution will pass by increasing parallelism in computer architecture. At least part of the solution will most likely come from a systematic use the relatively cheap GPUs for handling suitable algorithmic code, instead of the intrinsically serial CPUs.
 
Tasks (recommended length: 1000-3000 char):
This programme has strong components of both technology and particle physics, and the student will initially get academic training in the relevant subjects. The final goal of this PhD will be the implementation and real-life testing, of forefront jet substructure reconstruction methods as parallel algorithms running in GPUs. The student will follow the complete development cycle from algorithm conception to physics results using real ATLAS data, continuing the work of the Portuguese ATLAS team in LIP Lisbon.

The use of GPUs in scientific computing, and particularly in particle physics, is emerging, and promises to have an important impact. This PhD programme is therefore at the edge of both particle physics research and of computing technology, and the techniques developed will be extendable to other domains.
The student will integrate an international team working in the ATLAS experiment. He or she will take part in ATLAS data taking operations and physics analysis activities, with frequent trips to CERN to participate in Control Room shifts and collaboration meetings.
 
Research centre/lab or R&D unit hosting the thesis project:
Laboratório de Instrumentação e Física Experimental de Partículas
 
University to which the thesis project will be presented:
FC/UL - Faculdade de Ciências, Universidade de Lisboa
 
DAEPHYS Scientific Domain in which the project fits:
Instrumentation
 
Relation of the project to the Scientific Domains of DAEPHYS:
This project develops the use of advanced computing techniques and contemporary jet reconstruction advances necessary to reach the full potential of particle physics instrumentation and data acquisition in experiments.
 
Candidate profile:
The ideal candidate for this project is a motivated physics or engineering physics student, with good average marks, whose interests lie in the interface between the areas of computing and nuclear/particle physics.
 
Does this proposal involve more than one University?:
no
 
Synergies between the two Universities participating in the proposal:
DAEPHYS strongly encourages the presentation of thesis projects in co-supervision by researchers from two of the universities participating in the Program. In this field, explain the benefits resulting from the proposed co-supervision and the involvement of elements from the two universities, e.g. building critical mass teams, profiting from existing infrastructures or advanced equipments, profiting from expert technical know-how, etc. If the proposal involves only one University, write n/a.
(recommended length: 500-1000 char)
:
 
Does this proposal involve a company?:
no
 
Proposals involving a company:
DAEPHYS strongly encourages the presentation of thesis projects involving a company, preferably a high-tech company. These proposals have to: 1) be centered on a technological problem in which the partner company has been (or plan / would like to be) involved; 2) have a co-supervisor on the enterprise; 3) include part of the project to be carried out in the company.
(recommended length: 500-1000 char)
:

 

Supervisor

Name:Patricia Conde
Institution:Faculdade de Ciências da Universidade de Lisboa
email:pcmuino@ciencias.ulisboa.pt
 
link to CV or indication of ORCID ID:
0000-0002-9187-7478

 

Co-Supervisor

Name:Ricardo Goncalo
Institution:Laboratório de Instrumentação e Física Experimental de Partículas
email:jgoncalo@lip.pt
 
link to CV or indication of ORCID ID:
0000-0002-3826-3442

 

Uploaded PDF document: (none)