white paper

Low-Power Design Is a Corporate Mindset at Arm

SUCCESS STORY

Low-Power Design Is a Corporate Mindset at Arm

Arm® has been focused on providing scalable, low-power, high-performance IP since it was founded. The Arm architecture is the standard for low-power mobile applications, automotive/embedded solutions, and internet of things (IoT) designs. Most customers using Arm IP are interested in low-power architectures for their products. Therefore, low-power is a mindset at Arm, with every team member looking for power saving opportunities at every design level. From applications to the transistor level, Arm takes a system-level approach to low-power design.

The use of the PowerPro platform, outlined in this paper, provides ARM with an RTL design flow which is power-centric. The ability to perform daily RTL power analysis at the block/unit level provides rapid turnaround on the power trend, while weekly analysis provides more complete benchmark reference metrics.

Each element of the system can contribute to power consumption, but this paper focuses on designing power-efficient RTL IP, including:

  • Setting low-power objectives
  • Understanding the low-power design flow
  • Optimization techniques

Overall, the adoption of a unified RTL low-power methodology has helped Arm improve both RTL design efficiency and helped meet IP power goals by improving the PPA of their IP.

Share

Related resources

Xperi®: A Designer’s Life with HLS
Webinar

Xperi®: A Designer’s Life with HLS

This webinar will discuss two aspects of their experience going from RTL to HLS. The first topic is using HLS for algorithms such as Face Detection th

High-Level Synthesis Verification Technologies and Techniques
Webinar

High-Level Synthesis Verification Technologies and Techniques

This session will describe applying known and trusted static, formal and dynamic approaches to verification performed at the C++ or SystemC HLS level of abstraction.

NVIDIA: High-Level Synthesis in Agile System-on-Chip Flows: Overview and Techniques
Webinar

NVIDIA: High-Level Synthesis in Agile System-on-Chip Flows: Overview and Techniques

This talk provides a brief overview of NVIDIA Research’s use of Catapult HLS and highlights some useful features and flows of the Connections library, such as the ability to back-annotate SystemC simulations.