Memory Behavior Analysis for Accelerator-Chiplet Systems

Imec; Vlaams-Brabant; On-site; EUR 11,000 - 17,000; Full time

Benefits offered by this job

Internship allowance

Job summary

Imec in Belgium is seeking a Master internship to study memory behavior in accelerator-centric computing systems and chiplet-based architectures. The project analyzes how memory architecture influences performance, power consumption, and thermal behavior under realistic AI workloads.

The work integrates a DRAM simulator with an AI accelerator in a chiplet system to enable end-to-end memory-traffic analysis and to explore effects of workload characteristics and memory configurations on system

Qualifications

• Master student in electrotechnics, electrical engineering or computer science.
• Interest in memory architecture and its impact on performance.
• Familiarity with AI workloads is advantageous.

Responsibilities

• Study memory behavior under realistic AI workloads.
• Integrate a DRAM simulator with an AI accelerator in a chiplet system.
• Analyze impact on DRAM performance, power consumption and thermal behavior.
• Document findings and provide design-oriented insights.

Skills

Memory systems

Performance analysis

AI workloads

Education

Electrotechnics/Electrical Engineering

Computer Science

Tools

DRAM simulator

Chiplet systems

Power modeling

Job description

Imec in Belgium is seeking a Master internship to study memory behavior in accelerator-centric computing systems and chiplet-based architectures. The project analyzes how memory architecture influences performance, power consumption, and thermal behavior under realistic AI workloads.

The work integrates a DRAM simulator with an AI accelerator in a chiplet system to enable end-to-end memory-traffic analysis and to explore effects of workload characteristics and memory configurations on system