Postdoc Position on Design space exploration for Compute-Near-Memory architectures for advanced technologies 

What you will do

At imec you will explore the high-bandwidth, low-latency and low-power solutions provided by CNM architectures, analyzing common HPC workloads – Machine Learning and Augmented Reality - mapping them to CNM framework and heterogeneous systems. Your research will address: a) identifying workload patterns that naturally fit with the processing elements closer to the memory, exploring the tradeoffs of computing close to the caches and the main memory sub-system; b) build a framework to analyze the different compute near memory architectures for a given workload; c) propose new compute near memory architectures using SRAM, DRAM, MRAM, and other emerging memories for improving system performance.  

Research Area 

 Today’s memories subsystem organization is composed of hierarchical levels, from fast/small-sized multi-level caches to DRAM-based memories to slower/large storage class memory elements. Nowadays, workloads are highly constrained by the traditional dataflow across this memory hierarchy, which translates into enormous energy costs and performance penalties in every transaction involving memories from different modules within this memory ladder and compute units. Compute-near memory (CNM) addresses the data processing right next to the memory elements: by fusing memory with computing elements, thereby minimizing costly data movement. 3D technology solutions as face-to-face bonding and face-to-back TSV (through-silicon-via) enable memory and logic to be stacked together, and together with low-leakage and fast operation emerging non-volatile technologies, makes CNM a potential solution enabling future HPC architectures. 

What we do for you

We offer you the opportunity to join one of the world’s premier research centers in nanotechnology at its headquarters in Leuven, Belgium. You will be part of the STCO (System Technology Co-Optimization) program, working closely with a passionate group driving the future system architectures by connecting them to novel technologies. You will work closely with partners identifying the most relevant workloads and addressing the state-of-the-art challenges that industry leaders are facing. Being part of imec, your research will always be close to the cutting-edge memory, logic and interconnection technologies. With your talent, passion and expertise, you’ll become part of a team that makes the impossible possible. Together, we shape the technology that will determine the society of tomorrow.

We are committed to being an inclusive employer and proud of our open, multicultural, and informal working environment with ample possibilities to take initiative and show responsibility. We commit to supporting and guiding you in this process; not only with words but also with tangible actions. Through imec.academy, 'our corporate university', we actively invest in your development to further your technical and personal growth. 

We are aware that your valuable contribution makes imec a top player in its field. Your energy and commitment are therefore appreciated by means of a competitive scholarship. 

Who you are

• PhD degree in Electronic, Electrical Engineering or Computing/Science, Microelectronics 
• You have experience in design and benchmarking of computer architectures 
• You have experience in traditional memory hierarchies 
• You have experience in RTL and system modelling software frameworks 
• You are familiar with Compute-near-memory and Compute-in-memory topics
• It is important to enjoy working in an international environment, to be a good team player and to possess good written and oral communication skills in English.  

This postdoctoral position is funded by imec through KU Leuven. Because of the specific financing statute which targets international mobility for postdocs, only candidates who did not stay or work/study in Belgium for more than 24 months in the past 3 years can be considered for the position (short stays such as holiday, participation in conferences, etc. are not taken into account).