1 dag geleden - VU - Amsterdam
The Vrije Universiteit Amsterdam School of Business and Economics established the Hermine Weijland Fellowship in 2017 with the aim of attracting and promoting …
Eindhoven University of Technology has a vacancy for a PhD student on Architectures for In-Memory Computations in the Electronic Systems group, department of Electrical EngineeringProject descriptionThe classical Von Neumann architectures force data to be moved ...
Eindhoven University of Technology has a vacancy for a
PhD student on Architectures for In-Memory Computations
in the Electronic Systems group, department of Electrical Engineering
The classical Von Neumann architectures force data to be moved from a background memory to the processor before computations can be performed. Modern applications found in for example the domains of big data or advanced driver assistance do not fit well on this architecture. As a result, their performance is hampered by the memory bottleneck. Near memory computations have been proposed to partially address this bottleneck. The H2020 project MNEMOSENE takes this idea one step further. It will develop and demonstrate a new computation-in-memory (CIM) based architecture. This processor architecture will be based on memristors that combine storage and computation in a single device. To enable effective use of the architecture, the project will also deliver an accompanying programming flow and interfaces. It is expected that this new architecture will improve the energy-delay product by a factor of 1000X and the computational efficiency (#operations / total-energy) by a factor of 100X.
Within the larger MNEMOSE project, this PhD project will focus on the development of an innovative multi-tile memristor-based non-Von Neumann architecture and its CIM-ISA, which supports both storage and computation in the same physical location. The macro-architecture integrates many CIM tiles, as developed by other partners in the project. Since memristor tiles tightly integrate massive parallel computation with huge data storage, the traditional multi-level cache memory hierarchy has to be reconsidered, and will most likely be abandoned. This poses however new challenges for the compiler. Program and data locality must be exploited, as much as possible, directly within the tiles, and between the tiles. The PhD student will work closely with other academic partners to take these aspects into account within the MNEMOSE compiler. Finally, the PhD student will have to develop an instruction-set simulator for validation purposes and as a target for the compiler.
Departments and collaborators
Eindhoven University of Technology (TU/e) is a world-leading research university specializing in engineering science & technology. The Department of Electrical Engineering is responsible for research and education in Electrical Engineering. The discipline covers technologies and electrical phenomena involved in computer engineering, information processing, energy transfer and telecommunication. The department strives for societal relevance through an emphasis on the fields of smart sustainable systems, the connected world and care & cure. The TU/e is the world's best-performing research university in terms of research cooperation with industry (#1 since 2009).
The Electronic Systems (ES) group consists of six full professors, two associate professors, eight assistant professors, several postdocs, about 30 PDEng and PhD candidates and support staff. The ES group is world-renowned for its design automation and embedded systems research. It is our ambition to provide a scientific basis for design trajectories of electronic systems, ranging from digital circuits to cyber-physical systems. The trajectories are constructive and lead to high quality, cost-effective systems with predictable properties (functionality, timing, reliability, power dissipation, and cost). Design trajectories for applications that have strict real-time requirements and stringent power constraints are an explicit focus point of the group. Within this area, prof.dr. H. Corporaal and dr.ir. S. Stuijk have developed various novel power efficient computer architectures and their associated compilation trajectories.
The MNEMOSE project team is designed to combine extensive knowledge in the key fields (memristors, compilers and computer architectures). The project team includes industrial partners from ARM, IBM, and various leading universities in Europe. As part of this project, the PhD candidate will work closely with these industrial and academic partners to ensure that the developed architecture is suitable for the applications developed by these partners.
We are looking for candidates that match the following profile: