Researcher focusing on Time-Proportional Performance Analysis of GPUs

About the Job

We are seeking to hire a researcher in a temporary position for up to 15 months to work on instruction-level performance analysis approaches for Graphics Processing Units (GPUs). The position is affiliated with the DECAI project and builds on our prior work on instruction‑level performance analysis for out‑of‑order processors (featured at MICRO ’21, ISCA ’23, and ASPLOS ’26). These works produce various types of per‑instruction cycle stacks that explain what an application spends time on and why. Performance‑critical code is however increasingly run on GPUs, and the goal of this position is to build upon our prior work in the CPU domain to enable accurate, instruction‑level performance analysis on GPUs.

Your immediate leader is the head of the computing unit.

Duties of the Position

• Carry out top‑tier research focusing on instruction‑level performance analysis approaches for Graphics Processing Units (GPUs).
• Help DE...