NVIDIA is an American technology company that designs graphics processing units (GPUs) for the gaming market, as well as System-on-Chips (SOCs) for the mobile computing, data center and automotive market. NVIDIA Tegra System Software Performance Architecture Team (PAT) has cross-functional engagement with Tegra system architecture, core software and product teams to define platform architecture and performance requirements for key target markets. PAT establishes a foundation that ensures performance benchmarks and underlying architectural requirements are met for every generation of ARM SoCs. We developed a profiling methodology to help study workload and latencies within ARM Trusted Firmware (ATF), which is a software layer that enables various CPU and System power management functionalities in ARM SoCs. Our methodology allowed us to identify and analyze bottlenecks within ATF and provide insights into improving the efficiency of power management features in ATF.

• This report represents the work of one or more WPI undergraduate students submitted to the faculty as evidence of completion of a degree requirement. WPI routinely publishes these reports on its website without editorial or peer review.

Creator

• Jain, Adish
• Zimmerman, Ann

Publisher

• Worcester Polytechnic Institute

Identifier

• 115418
• E-project-121923-150913

Keyword

• Latencies
• Performance Benchmarking
• Embedded Systems
• Profiling
• NVIDIA
• Computer Architecture

Advisor

• Claypool, Mark L.

Year

• 2023

Sponsor

• NVIDIA

Date created

• 2023-12-19

Resource type

• Major Qualifying Project

Major

• Computer Science

Source

• E-project-121923-150913

Rights statement

• In Copyright

Last modified

• 2024-02-07