Enabling Cross-Domain Communication: How to Bridge the Gap between AI and HW Engineers
Authors:
Michael J. Klaiber,
Axel J. Acosta,
Ingo Feldner,
Falk Rehm
Abstract:
A key issue in system design is the lack of communication between hardware, software and domain expert. Recent research work shows progress in automatic HW/SW co-design flows of neural accelerators that seems to make this kind of communication obsolete. Most real-world systems, however, are a composition of multiple processing units, communication networks and memories. A HW/SW co-design process o…
▽ More
A key issue in system design is the lack of communication between hardware, software and domain expert. Recent research work shows progress in automatic HW/SW co-design flows of neural accelerators that seems to make this kind of communication obsolete. Most real-world systems, however, are a composition of multiple processing units, communication networks and memories. A HW/SW co-design process of (reconfigurable) neural accelerators, therefore, is an important sub-problem towards a common co-design methodology. The ultimate challenge is to define the constraints for the design space exploration on system level - a task which requires deep knowledge and understanding of hardware architectures, map** of workloads onto hardware and the application domain, e.g. artificial intelligence.
For most projects, these skills are distributed among several people or even different teams which is one of the major reasons why there is no established end-to-end development methodology for digital systems. This position paper discusses possibilities how to establish such a methodology for systems that include (reconfigurable) dedicated accelerators and outlines the central role that languages and tools play in the process.
△ Less
Submitted 8 April, 2021;
originally announced April 2021.
