From Performance Engineering Lab


Lero MSc (by Research) and PhD positions 2017

Research Project Area

Performance Management and Testing of Large Scale Systems

Description

The Performance Engineering Laboratory (PEL) is one of the key research laboratories within the School of Computer Science at University College Dublin. Its main goal is to focus on the design, simulation and demonstration of new technologies for future enterprise-level software and telecommunication systems.

PEL has openings for several ambitious postgraduate students (at both MSc and PhD level) to work on a set of collaborative projects with industrial partners, such as IBM Research and IBM’s Software Laboratory in Dublin. This is an exciting opportunity to gain real-world industrial software engineering experience, while also studying for a research-based MSc/PhD in the School of Computer Science at UCD. These projects will focus on state-of-the-art research topics related to Software Performance Engineering. The postgraduate students will be conducting world-class research work, including the experimental evaluation of the same. This might involve implementing tools or prototypes based on the discussions made with engineers and researchers based in UCD as well as our industrial partners, as well as evaluating the feasibility and effectiveness of the proposed solutions. Examples (that do not necessarily reflect the exact topic of the postgraduate degree, as it which will be agreed between the academic and industrial partners when the students are in place) of research projects jointly done by PEL and our industrial partners are:

  • Developing data analytics tools to automatically monitor and detect performance problems in large scale enterprise systems (e.g., cloud-based) or even for software-defined data centres (i.e., when the amount of activities logged is growing exponentially).
  • Assessing different strategies that can be used to improve the performance of Hadoop Distributed File System solutions, so that they are most suitable for Big Data scenarios.
  • Developing techniques to dynamically adapt the workload used by a performance testing tool in order to improve the process of identifying workload-dependent performance issues, as well as their root causes, in highly distributed environments, such as The Cloud.
  • Investigating log analytics techniques with embedded expert knowledge in order to improve the inference and identification of anomalies in Internet-of-Thing systems (e.g., a Smart building).
  • Efficiently distinguishing between meaningful and noisy events in a stream-based log dataset. In particular, the aim is to minimise false alarms by accurately determining if a given event reflects a ‘real’ abnormal system behaviour that requires intervention from a system administrator.
  • Identifying the root cause of anomalies detected in large volumes of system log data. There are plenty of techniques available (e.g., using rules, models or patterns), but they do not always work well at the scale envisaged, and some re-coding of existing and novel techniques over the distributed environment will be required.

Finally, these MSc/PhD positions will be based in the Performance Engineering Lab at the UCD School of Computer Science, with some significant time spent in the corresponding industrial partner’s premises.

Required Skills Profile

The successful candidate should be self-motivated with the enthusiasm to develop technical skills across a range of disciplines. Moreover, he/she must have:

  • An honours degree in Computer Science (or a closely related field).
  • Excellent software programming skills.
  • Excellent written, communication and presentation skills.
  • Experience in Java development (or similar programming language, such as Python) is desirable.
  • Strong sense of ownership and drive.
  • Applicants whose first language is not English must provide evidence of competency in English.


Environment

The Performance Engineering Laboratory (PEL) combines engineering research in the areas of software, security and privacy, as well as data networks – in short, anything where performance issues might arise and where the application of performance analysis can support the identification and resolution of those issues. PEL brings together researchers from different professional and cultural backgrounds. People in PEL come from a number of different countries: Ireland, France, Italy, Algeria, Brazil, Mexico, China and more. Expert knowledge varies from post-graduate to experienced senior researchers and academic staff.

Conditions and Benefits

These 2-year MSc and 4-year PhD positions are fully funded by Lero, the Irish Software Research Centre (http://www.lero.ie/) through Science Foundation Ireland. The successful candidates will receive a tax-free salary stipend and full coverage of their UCD’s fees. UCD student status gives access to UCD facilities and services (such as clubs, concerts, sports centre, campus pubs, etc.). UCD is the leading Irish University and gives access to many other on-campus facilities.

How to Apply

These positions are immediately available. The Principal Investigator on these projects is Professor John Murphy, and applications are accepted anytime from now until the positions are filled.

Interested applicants should submit: (1) CV, (2) academic transcripts of studied degrees, and (3) a covering letter explaining the interest in the position (including if the applicant is interested in either a MSc or Ph.D. postgraduate degree) to Dr. Omar Portillo () who is managing this recruitment process. Finally, please indicate "PEL postgraduate position 2017" in the subject of your e-mail.