Aeonic Generate Digital PLL for multi-instance, core logic clocking
Industry Expert Blogs
Simplifying workload modeling with AMBA ATP Enginearm Blog - Francisco Socal, ArmMay. 25, 2020 |
We are pleased to announce the release of the AMBA ATP Engine, an open-source implementation of AMBA ATP (Adaptive Traffic Profiles). The Engine significantly simplifies the adoption of AMBA ATP for workload modeling and accelerates the research and development of heterogeneous systems.
AMBA ATP and the need for a workload modeling framework
Computing systems are becoming increasingly complex and computationally demanding. Heterogeneous systems with accelerators and multiple computing cores are used to enable new workloads efficiently and to address increased performance requirements.
This creates challenges for workload modeling, system performance analysis, and architecture exploration of systems-on-a-chip (SoCs).
With that in mind, we developed AMBA Adaptive Traffic Profiles, or AMBA ATP. The AMBA ATP Specification defines a workload modeling framework that balances complex synthetic workload generation with ease-of-use.
AMBA ATP manages complexity by modeling the high-level memory access behavior in a concise, simple, and portable way. Traffic profiles can be used across multiple tools, design, and verification environments to assist with the design and verification of complex SoCs. Traffic profiles enable a simpler and faster simulation mechanism that is simultaneously predictable and adaptive.
AMBA ATP is gaining momentum and is adopted by our ecosystem partners in commercial products and research projects.