This role pays more than
90%
of similar roles. Most pay
$184,168–$264,875
— the shaded band above.
At the midpoint, this role pays about
$309k
versus about
$225k
for comparable roles.
Based on 240 similar postings.
Employer
About Arm Holdings
Arm Holdings plc is a leading British semiconductor and software design firm, established in 1990 and recognized for developing energy-efficient processor architectures that power nearly all smartphones and a vast range of IoT and computing devices.
Arm Holdings currently has
109 open roles
on FindRole.
Listed pay typically runs
$209,100–$282,900
across 108 roles with salary data.
As a Principal SoC Performance Architect – Next-Generation Server Platforms, you will join a team focused on pushing performance boundaries in modern server systems through architecture, workload characterization, and silicon-level analysis. You will be responsible for evaluating real-world workloads on production silicon, specifically targeting AI at scale and HPC applications. Your daily work involves partnering with multi-functional teams to define performance metrics, designing targeted microbenchmarks to isolate architectural behaviors, and identifying bottlenecks to recommend optimizations. You will translate insights into architectural improvements for scalability while providing technical leadership across hardware and software teams. The role requires expertise in CPU/accelerator architectures, cache and memory subsystems, and high-performance networking technologies like PCIe, CXL, InfiniBand, and RoCE. Required technical skills include proficiency in SystemC, TLM, C++, Python, and experience with performance analysis tools or simulation environments.
Analyze real-world workloads on production silicon with a focus on AI at scale and HPC applications.
Define performance metrics, representative use cases, and benchmarking strategies for future platforms.
Design and implement targeted microbenchmarks to isolate and study specific architectural behaviors.
Identify performance bottlenecks through detailed analysis and recommend actionable optimizations.
Translate insights from workload and silicon studies into architectural improvements for performance and scalability.
Provide technical leadership as a performance expert across multiple hardware and software teams.
Drive end-to-end understanding from architecture to workload to silicon validation to ensure optimal system performance.
What we're looking for
Master's or PhD degree in Computer Engineering, Computer Science, or a related field.
10+ years of experience in system or hardware architecture including CPU/accelerator architectures, cache and memory subsystems, PCIe, networking, or storage.
Strong background in high-performance IO/networking or performance optimization for HPC or cloud-scale systems.
Proven expertise in benchmarking methodologies and the development of custom microbenchmarks.
Solid experience with performance analysis tools, profiling, and simulation environments.
Proficiency in SystemC, TLM, C++, and Python.
Ability to navigate and contribute effectively within large, complex codebases.
Experience with silicon bring-up or post-silicon performance analysis (preferred); familiarity with performance simulators or architectural modeling tools (preferred).
Remote (Santa Clara, CA)+316 days ago
$209,300–$313,900
SoC Architecture
CPU Microarchitecture
Performance Modeling
Linux
Firmware
Memory Management
PMU
Data Analysis
Scripting
System Configuration
I/O Subsystems
C++
Python
Lua
SoC Architecture
Machine Learning
Data Science
RTL
Shell Scripting
STL
Queuing Theory
Memory Systems
QoS
Arbitration
Data Analytics
Statistics