P99 Conf Logo
SESSION ON-DEMAND

All Things P99

The event for developers who care about P99 percentiles and high-performance, low-latency applications

Back to All P99 CONF Sessions
More Performance Strategies

Unlocking Code-level Performance Analysis with PMUv3 Plugin

Profiling code sections on ARM platforms is hard with whole-application tools. This talk introduces the PMUv3 plugin, which automates metric collection and visualization at the code block level using ARM’s PMUv3 library. It supports 70 hardware events and 40 KPIs across 15 bundles, with APIs to target specific functions or workloads and scripts to generate comparative graphs. Applied to 5G codebases like srsRAN and Radisys, the plugin cut profiling time, empowered ARM teams to analyze workloads independently, and provided actionable insights for performance tuning.

19 minutes
Register for access to all 60+ sessions available on demand.
Fill out the form to watch this session from the P99 CONF 2025 livestream. You’ll also get access to all available recordings.

Gayathri Narayana Yegna Narayanan, Senior Solutions Engineer at Arm, Inc.

As a 5G Senior Solutions Engineer at ARM, I specialize in orchestrating the stack bring-up process and conducting in-depth code analysis to optimize system performance. My focus involves integrating 5G components, such as RAN and Core Network, with precise configurations to ensure seamless operation. I delve into critical code components, analyzing their impact on system and CPU performance to identify areas for enhancement. Additionally, I play an active role in developing software plugins and tools for studying 5G performance across different generations of ARM processors. My goal is to continually improve code execution efficiency and overall system performance. Prior to this, I was a Graduate thesis student at Texas A&M University with majors in Computer Engineering, where I was actively involved in the research of 5G Networks using Reinforcement Learning associated with the LENS Laboratory.