Job Detail

Ref. 2018-38

About the position

We are seeking an outstanding PhD student to pursue a challenging research project in the field of computer system design for big data analytics and artificial intelligence. The position has a duration of four years. The PhD student will work in the Cloud Storage and Analytics group at IBM Research – Zurich.

Project

Many search and combinatorial optimization problems involve exploration of a large solution space to identify possible solutions to optimize an objective function under given constraints. For instance, in the text and graph-mining fields, finding patterns that exhibit certain properties is typically of high interest. In the case of computer games, such as Go, the search involves finding a set of valid moves that can lead to an advantage and eventual victory. When formulated recursively, such search programs can be very concise. However, the resulting computation is typically very irregular and exhibits limited single-thread performance and SIMD (single-instruction multiple-data) parallelism. One of the objectives of this project is to develop the necessary software and hardware infrastructure to enable scalable implementations of recursive search and optimization algorithms on massively parallel compute clouds.

In the course of this project, we are planning to target massive-scale graph analytics problems, e.g., the Graph Challenge recently created by DARPA and MIT, which involves detecting communities in social networks with billions to trillions of nodes and edges using subgraph-isomorphism and hierarchical-stream-clustering techniques. Such large data sets inherently do not fit into a single storage or memory device and require efficient mechanisms for retrieving information from a distributed storage, e.g., by means of an index. In addition, forming clusters within this data requires the availability of similarity information. Therefore, a secondary objective of this project is to develop efficient information-retrieval mechanisms based on scalable similarity search and indexing techniques.

Requirements

Candidates are expected to have the following background and interests

• M.Sc. degree in Computer Engineering, Electrical Engineering, or Computer Science


• A strong background in parallel computer architectures and parallel programming


• A strong understanding of graph algorithms or machine-learning techniques


• Excellent software skills and familiarity with functional programming


• Strong communication skills (both written and verbal)


• Motivation to acquire new skills and ability to think out-of-the-box

Would be a plus

• Experience in programming FPGAs, GPUs, and/or many-core processors


• Experience in hardware description languages and event-driven simulation


• Familiarity with cloud computing, virtual machines, and containers

Diversity

IBM is committed to diversity at the workplace. With us you will find an open, multicultural environment. Excellent flexible working arrangements enable both women and men to strike the desired balance between their professional development and their personal lives.

How to apply

Please send your application documents including CV, publication list, and a brief cover letter explaining your motivation in a single PDF file. Reference letters are welcome. The position is available immediately.

Dr. Kubilay Atasu
Research Staff Member
IBM Research – Zurich