Energy-Aware HPC/AI Performance Engineer

Inria Talence; Talence; On-site; EUR 45,000 - 65,000; Full time

Job summary

Inria Talence is seeking a researcher to design and implement a methodology and an open-source toolkit for collecting, storing and processing HPC/AI traces, focusing on energy consumption, I/O behavior, and resource utilization. You will work with BSC, CINECA and Latin American partners to deploy the toolkit across European and LAC HPC centers.

The role involves instrumenting executions, developing dashboards, and drafting a methodology guide including GDPR-aligned data governance, with the aim

Qualifications

• Experience with HPC profiling and energy measurement.
• Familiarity with tracing tools and dashboards is a plus.

Responsibilities

• Design and implement a methodology for collecting HPC/AI traces.
• Develop an open-source toolkit for tracing, energy measurement and dashboards.
• Instrument HPC workflows and provide performance and energy dashboards.

Skills

C/C++

Python

Bash

Unix CLI

Slurm

Profiling tools

RAPL

NVML

Grafana

Prometheus

Tools

Darshan

Tau

Recorder

IOPS

Prometheus

InfluxDB

Job description

Inria Talence is seeking a researcher to design and implement a methodology and an open-source toolkit for collecting, storing and processing HPC/AI traces, focusing on energy consumption, I/O behavior, and resource utilization. You will work with BSC, CINECA and Latin American partners to deploy the toolkit across European and LAC HPC centers.

The role involves instrumenting executions, developing dashboards, and drafting a methodology guide including GDPR-aligned data governance, with the aim