The Rest Is History

ESL has been around long enough that it’s time to examine some of the historical assumptions.


I recently fielded some questions on the evolution of Electronic System Level design. The questions started me thinking about how ESL is being applied today and what effect some of the historical assumptions might be having on ESL’s perceived use cases.

It’s interesting that we are at a point to consider historical assumptions about ESL. Looking it up, Wikipedia claims that the term ESL Design was first defined in February 2001. I guess after 12 years—starting when the leading-edge process technology was at 130nm—there may need to be some evolution in the use model.

In thinking about this I realized that in practice ESL has not been limited to only the “Electronic System Level.” ESL has been applied to the complete system design. ESL captures important characteristics of the system, how those characteristics are implemented at a functional and performance level, and how the elements that implement those characterizations interact.

There are many cases in which critical aspects of the system are not purely electronic. In ESL models we can capture virtually any aspect of the system, and implement that characteristic in a transaction-level model. To be sure, the bulk of the system from a modeling perspective may be electronic, but the key interactions that provide the most value in System-Level analysis often incorporate aspects of the interaction of the electronic portions of the system with the non-electronic portions of the system.

From my experience those using ESL design techniques to model and analyze their systems are not restricting themselves to the electronic portions of the system. They are in fact modeling the interaction of critical system components at a level appropriate to the type of questions and analysis required. I’ve seen many instances where this includes electro-mechanical interaction with motors and sensors, and even cases in which biological structures, such as a heart, have been modeled along with electronic elements in an ESL design.

I believe that in practice, users of ESL design techniques are treating the “Electronic” aspect of ESL as a suggestion. The “E” in ESL is important as an historical starting point, but the use of ESL has evolved to include appropriate representations and interactions with other domains. Users have applied the ESL techniques to address their needs. After all isn’t an aspect of evolution, the gradual transformation of a successful system to more successfully satisfy the demands placed on the system?