PhD in Efficient Memory System for Infinite-Dimensional Compute Eindhoven University of Technology in Netherlands

Are you passionate about overcoming the memory wall and designing next-generation memory systems for complex data? This PhD position at Eindhoven University of Technology (TU/e) offers the opportunity to explore new architectures, addressing schemes, and optimizations that combine hardware and software insights to build scalable, energy-efficient solutions for data-intensive applications. Modern high-performance applications, from scientific simulations to machine learning workloads, are increasingly limited by the performance and energy constraints of current memory systems. As data becomes more complex and high-dimensional, efficiently storing, accessing, and processing this information is a critical bottleneck.

This project focuses on designing next-generation memory systems for efficient processing of complex, multidimensional data representations. You will develop simulation frameworks to model application behavior on memory technologies such as DRAM and 3D-stacked memory, enabling detailed analysis of access patterns, bandwidth requirements, and latency constraints. These insights will guide the identification of key performance bottlenecks and opportunities for optimization. Building on this understanding, you will design novel memory architectures and optimization techniques, including efficient address-mapping strategies, hardware prefetching, adaptive precision encoding, and memory-aware scheduling. You will also explore in-memory and near-memory computing approaches to reduce costly data movement and improve overall system efficiency. These innovations aim to enable scalable and predictable performance for data-intensive and irregular workloads.

Additionally, you will investigate emerging memory technologies such as optical-electrical memory (OEM), which offer significant opportunities for improved energy efficiency and performance. The research is conducted in close collaboration with experts across computer architecture, software systems, and device technology, contributing to a holistic, memory-aware computing framework that bridges the gap between application requirements and next-generation memory system capabilities.

The position is embedded in the Electronic Systems (ES) group within the Electrical Engineering department at TU/e. The ES group is world-renowned for its design automation and embedded systems research, consisting of six full professors, four associate professors, six assistant professors, several postdocs, and approximately 40 EngD and PhD candidates. The group provides a vibrant and collaborative research environment.

Interested candidates should apply online via the AcademicTransfer portal. For further information, contact the Electronic Systems group at TU/e.