[[Apologies for multiple copies. |
Please forward this message to any potentially interested candidate that you may know.
The intended starting date (1st of December, 2011) is very close, but excellent students that are currently working on their Master's thesis and will finish within a couple of months, are also very welcome to apply.
Interested candidates can always contact me for further information.
I'm looking forward to having many reactions.
Thanks to you all for your help!
University of Twente]]
The research group Formal Methods and Tools at the University of
Twente (Enschede - The Netherlands) is looking for
a PhD researcher (4 years)
to work on the EU Strep project CARP (Correct and Efficient Accelerator Programming), funded by the European Union. Other partners in the project are Imperial College, UK; Realeyes, Estonia; ARM, UK; RTWH Aachen, Germany, Monoidics, UK, ENS, France and Rightware, Finland.
Within the context of the CARP project, the PhD student will work
in particular on:
- requirements analysis for accelerator programming
- the formal semantics of an intermediate programming language for describing accelerator algorithms
- developing logic-based verification techniques for this intermediate programming language, taking into account common accelerator programming patterns
In recent years, massively parallel accelerator processors, primarily GPUs, have become widely available to end-users. Accelerators offer tremendous compute power at a low cost, and tasks such as media processing, simulation, medical imaging and eye-tracking can be accelerated to beat CPU performance by orders of magnitude. Performance is gained in energy efficiency and execution speed, allowing intensive media processing software to run in low-power consumer devices.
Accelerators present a serious challenge for software developers. A system may contain one or more of the plethora of accelerators on the market, with many more products anticipated in the immediate future. Applications must exhibit portable correctness, operating correctly on any configuration of accelerators, and portable performance, exploiting processing power and energy efficiency offered by a wide range of devices.
The overall aims of CARP are to design techniques and tools for correct and efficient accelerator programming:
- Novel & attractive methods for constructing system-independent accelerator programs
- Advanced code generation techniques to produce highly optimised system-specific code from system-independent programs
- Scalable static techniques for analysing system-independent and system-specific accelerator programs, both qualitatively and quantitatively
The PhD candidate we are looking for is expected to work on the development of tools and techniques for correct accelerator programming. Within the consortium an intermediate programming language for accelerator programming will be developed. The PhD candidate is expected to develop a solid formal semantics for this language, together with appropriate verification techniques. An important focus of the verification work is that it will focus on bug finding, without too many false negatives, rather than on developing full correctness proofs. For the verification, it is expected that the logical basis will be permission-based concurrent separation logic. Sometimes the intermediate programming language will not provide the required efficiency, and programs will be written in a low-level language as OpenCL. Therefore, the verification techniques also will be extended to this lower level.
An enthusiastic PhD student with an MSc degree in Computer Science (or an equivalent qualification).
The candidate should have
- a thorough theoretical background,
- a demonstrable interest in program semantics and verification, and
- some knowledge about multithreaded programming (in Java/C/C++).
We are looking for a researcher with an independent mind who is willing to cooperate in our team. It is understood that he or she works on the topics listed above, and contributes to the expected deliverables for the project. Further we ask for good communicative and good collaboration skills. Candidates should be prepared to prove their English language skills.
As a research outcome we expect publications, (prototype) tools, and a PhD thesis.
Starting date of the position: December 1st, 2011, or as soon as possible thereafter.
- A PhD position for four years (38 hrs/week)
- A stimulating scientific environment
- Gross salary ranging from EUR 2042 tot E 2612 (4th yr) per month
- Holiday allowance (8%), end-of-year bonus (8.3%)
- Good secondary conditions, in accordance with the collective labour agreement CAO-NU for Dutch universities
- A green Campus with lots of sports facilities
You will be a member of the Twente Graduate School in the research programme 'Dependable and Secure Computing' under the leadership of Prof Dr Jaco van de Pol. The research programme offers advanced courses to deepen your scientific knowledge in preparation to your future career (within or outside academia). We provide our PhD students with excellent opportunities to broaden their personal knowledge and to professionalise their academic skills. Participation in national and/or international summer schools and workshops, and visits to other prestigious research institutes and universities can be part of this programme.
- FMT group: http://fmt.cs.utwente.nl/
- Dr. Marieke Huisman (Marieke.Huisman@ewi.utwente.nl)
- The CARP project: http://fmt.ewi.utwente.nl/files/projects/CARP.d1.pdf
Please submit your application before 15th of November, 2011 via http://www.utwente.nl/vacatures/en/. We strongly encourage interested applicants to send in their applications as soon as possible.
Your application should consist of:
- a cover letter (explain your specific interest and qualifications);
- a full Curriculum Vitae, including a list of all courses + marks, and a short description of your MSc thesis; and
- references (contact information) of two scientific staff members.