Although SystemVerilog and VHDL provide the capabilities to describe algorithms and decision trees at a high level of abstraction they are fundamentally hardware centric. SystemC, probably the most used language to describe hardware systems at a level above register transfer, also has its limitations, especially in the temporal domain. Systems architects need to describe the requirements and architecture of a system without the limits imposed by a presumed implementation choice. The combination of MATLAB and Simulink comes close to fulfilling these requirements, but it is a proprietary solution and thus cannot be seen as a viable alternative to the creation of a new standard. I personally know of two languages that either already are being used for system design or offer the possibility of such application. And a third language is in its final stages of development. Rosetta as this language is called, in fact, would have been available much sooner had it received the funding it deserved. You can find more information about the language and its authors on this web page. One of the two existing languages is more popular in Europe than in the US. The B language, not to be confused with another B language developed at Bell Labs, is a formal language for the description of systems. To find out more about it click here. It helps to know a bit of French in navigating this site, but there is enough English to get you to the appropriate information. Finally, the third method that is finding supporters for the description of systems is XML. Belonging to the same family of languages at HTML, and SGML, XML, for extensible Markup Language, specifies the format you can use to describe data, but leaves totally open the nature of the data. The article is an IBM publication and provides a good overview of the characteristics and limitation of this standard.
To comment on this article or to offer further input please go to the Forum section of this site under the ESL Talk topic.
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.