GPU Power Efficiency Engineer

Summary

Posted: Nov 8, 2024

Role Number:200578296

Do you love creating elegant solutions to highly complex challenges? As part of our Silicon Technologies group, you’ll help design and manufacture our next-generation, high-performance, power-efficient processor, system-on-chip (SoC)! You’ll ensure Apple products and services can seamlessly and efficiently handle the tasks that make them beloved by millions. Joining this group means you’ll be responsible for crafting and building the technology that fuels Apple’s devices. Join us to help deliver the next groundbreaking product containing an Apple designed GPU! In this highly impactful role, you will be responsible for Efficiency or Performance per watt projections for the GPU from early feature specification stages to Silicon. As a member of the Silicon Technologies GPU team, you will work across functions including Architecture, Power, Performance, Silicon Validation, Thermals and Technology. The job involves analyzing and influencing key factors effecting Performance and Power of the GPU and improving on methodologies to align the model to Silicon.

Description

Join the team to play a pivotal role in Performance and Power projections of the GPU. As Apple expands on its GPU feature set, we rely on making accurate Efficiency curves for Performance per watt. You will have the opportunity to influence GPU Architecture choices and the feature set for the product. You will work with many teams including Architecture, Power, Performance, Silicon Validation, Thermals and Technology. In the highly collaborative GPU team, your main responsibilities are: – Responsible to generate and analyze GPU Efficiency Projections, Performance per watt, for each generation of GPU over the full cycle of the project, from concept to Silicon. – Drive the methodology of generating projections to improve accuracy and quality. – Responsible for System and Chip level Power team requests to the GPU for projections. – Work with Technology, Power and Static timing teams to determine the Frequency-Voltage curves for the workloads. – Work with Performance and Power teams to gather data for the Efficiency curves and analyze the impact. – Work with the Silicon Validation team to improve projections based on Silicon data. – Collaborate with GPU architecture team to set power targets and determine the impact of the features on efficiency.

Minimum Qualifications

• BS + 10 years relevant experience
• Experience in areas of Power and Silicon technology
• Experience with Python, Perl, or other scripting/coding skills
• Experience in data analysis, data mining and triage

Preferred Qualifications

• Knowledge of low power design and implementation techniques
• Strong communication skills and ability to work with cross-functional groups
• Experience in ASIC power analysis in PowerArtist, PTPX and power projections
• Experience in post-silicon power correlation
• Experience in power management techniques
• Familiarity with Performance modeling
• Experience with GPU workloads is a plus