Exploration of Compute-near-Memory Processing Unit Architectures for High Performance Systems
Master projects/internships - Leuven | More than two weeks ago
Examining novel near-memory accelerator architectures for modern applications and memory models.
Systems that deploy computational logic near memory can overcome typical von Neumann-based bottlenecks (e.g. memory wall) by limiting the amount of data transferred to central compute areas in a system. With modern compute-near memory (CnM) systems still in their infancy and typical programming paradigms focused on centralized computing however, new work must be undertaken to understand the power, performance, and area trade-offs of specific CnM-instruction set architecture (ISA) implementations.
In this internship, you will learn how to use imec’s system-level framework for examining novel memory technologies and compute-near-memory (CnM) processing cores. You will then implement and evaluate new capabilities for the CnM hardware model for a variety of application scenarios.
Key responsibilities will include:
- Learning and using DRAM simulators to model memory technologies and protocols and perform performance analyses.
- Performing power and area analysis using Catapult HLS tools.
- Examining new CnM hardware extensions for new applications.
This role is ideal for someone who is deeply interested in hardware-software codesign, electrical engineering, and working in an interdisciplinary environment that values innovation, creativity, and real-world impact.
Profile: You are analytical and detail-oriented, with a strong interest in memory technologies and near-memory accelerators, and their impact on high-level applications such as AI, genomics, databases, forecasting, or other datacentre-class algorithms. You are adept at or have a keen interest in high-level synthesis, programming, and modelling.
Background: Computer engineering, electrical engineering, memory, accelerators. Knowledge of SystemC, Catapult tools, and/or scripting languages is an advantage.
Master's degree: Master of Science; Master of Engineering Technology; Master of Engineering Science
Master program: Computer Science; Electrotechnics/Electrical Engineering
Type of Project: Thesis; Internship; Combination of internship and thesis
Duration: 6-9 months, possibility to extend further.
For more information or application, please contact Joshua Klein (joshua.klein@imec.be) and Leandro M. Giacomini Rocha (leandro.m.giacominirocha@imec.be).
Imec allowance will be provided.