Logo of Huzzle

Technical Intern, CPU Architecture

image

Ampere

Jan 19, 2024

Applications are closed

  • Internship
    Full-time
    Off-cycle Internship
  • Architecture & Urban Planning
  • Portland

Requirements

  • Candidate must be pursuing MS in Computer Science, Computer Engineering, or Electrical Engineering. PhD is a plus.
  • Strong skills in C++ and/or SystemC
  • Strong skills in Python or other scripting languages
  • Experience with automation, scripting, and performance tools
  • Experience with performance benchmarks: analysis and optimization
  • Experience developing and correlating performance/simulation models
  • Experience in Excel, including Visual Basic, Pivot Tables, Macros
  • Experience in R statistical package
  • Experience using hardware performance counters.
  • Basic understanding of computer systems performance principles, e.g. Amdahl’s Law, Queuing Theory, mean value analysis. Great communications skills
  • Understanding of CPU microarchitecture
  • Comfortable in using a Linux development environment
  • Excellent troubleshooting and problem-solving skills
  • Internship is full-time for Summer or Fall in Portland, OR or Santa Clara, CA.

Responsibilities

  • Learning the CPU performance model from senior team members and helping to develop, deploy, and run microarchitecture studies for future products.
  • Working with microarchitecture designers in performing studies on current and future generation of CPUs.
  • Driving performance correlation of models using post-silicon measurement data or pre-silicon simulation results.
  • Collecting silicon performance measurements and understanding how to vet data and filter anomalous results.
  • Developing automated performance testing and data post-processing.
  • Developing visualization tools to assist in performance analysis.
  • Running micro-benchmarks and creating supporting models/relationships using the collected data.
  • Developing spreadsheet macros to further prepare/present performance data in appropriate formats.
  • Investigating performance anomalies with help from senior performance architects.
  • Developing functional/performance model of the hardware in C++ or System C.
  • Fitting curves and running trendline analysis using R.

Delivering the New Standard for Cloud and Edge Servers.

Manufacturing & Electronics
Industry
1001-5000
Employees
2017
Founded Year

Mission & Purpose

Ampere is a modern semiconductor company designing the future of hyperscale cloud and edge computing with the world’s first Cloud Native Processors. Built for the sustainable cloud with a modern 64-bit Arm server-based architecture, Ampere works with leading cloud service providers to enable them to accelerate the delivery of all cloud computing applications. With industry-leading cloud performance, power efficiency and scalability, Ampere processors are tailored for the continued growth of cloud and edge computing.