IBM Research – Zurich and ETH Zurich are seeking an outstanding PhD student to pursue a challenging research project in the field of computer system design for Big Data analytics.
The position is available immediately. The duration of the project is four years. The project will be supervised by Dr. Kubilay Atasu (IBM Research) and Prof. Onur Mutlu (ETH).
The successful candidate will enjoy a competitive salary and work in an international research environment.
Many Big Data analytics problems can be formulated and solved using recursive programs. Although recursive programs are concise, they are hard to parallelize and exhibit limited single-thread performance and single-instruction multiple-data parallelism. Computer architectures that are specifically designed to support parallel recursion can overcome the performance limitations of general-purpose architectures and achieve the energy efficiency required by datacentres and mobile devices. Supporting parallel recursion at scale requires 1) a large number of processing cores that can operate in parallel on independent regions of input data structures; 2) a domain-specific network that interconnects the processing cores and the memory system, and 3) a dynamic scheduling logic that maximizes the utilization of the available processing cores. This research project will address fundamental questions, such as the following:
– What are the capability and the number of processing cores needed per chip to meet performance goals?
– Is there a scalable memory system architecture that can service all processing cores in parallel?
– What network architecture is needed to enable dynamic scheduling and to achieve maximal utilization?
– How can we virtualize the resulting architecture and enable multiple users and applications to share it?
– MSc degree in computer engineering, electrical engineering, or computer science
– A strong background in parallel computer architectures and parallel programming
– Excellent software skills in C/C++ or Java
– Familiarity with graph algorithms or machine learning techniques
– Experience in hardware description languages, such as Verilog, VHDL, or SystemC
– Experience in programming FPGAs, GPUs or many-/multi-core processors
– Experience in cloud computing platforms, virtual machines, and containers
– Experience in designing memory or storage systems and controllers
Applicants should send their CVs and a brief research statement to Dr. Kubilay Atasu (email@example.com).