TY - GEN
T1 - Introducing Model-of-Things (MoT) and Model-of-Design (MoD) for simpler and more efficient hardware generators
AU - Ecker, Wolfgang
AU - Schreiner, Johannes
N1 - Publisher Copyright:
© 2016 IEEE.
PY - 2016/11/22
Y1 - 2016/11/22
N2 - Several leading research groups name hardware generation as the next disruptive productivity improvement after IP-reuse. Metamodeling and code generation have already demonstrated a speedup by a factor 3× for the complete implementation phase of a chip. Furthermore, code size reduction by a factor of 3× was achieved with the hardware generation language (HGL) Chisel. However, the development of generators, often implemented as scripts with print statements or with template engines, takes a significant amount of time. As a countermeasure, the OMG vision of Model Driven Architecture (MDA) introduces intermediate models and transformations between them to structure the generation process and to provide a measure for reusing parts of the generators. In this paper, we present our adoption of MDA to digital hardware design to speedup generator construction. We define intermediate models named Model-of-Things (MoT) for formalizing specification data and Model-of-Design (MoD) for hardware design related modeling. In addition, we introduce Templatesof- Design (ToD) to capture the (micro-)architecture, to retrieve MoT data from the MoD, and to enable further automation in specifying the architecture. Finally, we define basic type and expression models that are used in MoT, MoD as well as ToD and thus simplify passing data from one model to another. First analysis shows that we can simplify building one generator up to a factor of 10x. This factor increases further when different target languages or target language styles are generated from one MoD as well.
AB - Several leading research groups name hardware generation as the next disruptive productivity improvement after IP-reuse. Metamodeling and code generation have already demonstrated a speedup by a factor 3× for the complete implementation phase of a chip. Furthermore, code size reduction by a factor of 3× was achieved with the hardware generation language (HGL) Chisel. However, the development of generators, often implemented as scripts with print statements or with template engines, takes a significant amount of time. As a countermeasure, the OMG vision of Model Driven Architecture (MDA) introduces intermediate models and transformations between them to structure the generation process and to provide a measure for reusing parts of the generators. In this paper, we present our adoption of MDA to digital hardware design to speedup generator construction. We define intermediate models named Model-of-Things (MoT) for formalizing specification data and Model-of-Design (MoD) for hardware design related modeling. In addition, we introduce Templatesof- Design (ToD) to capture the (micro-)architecture, to retrieve MoT data from the MoD, and to enable further automation in specifying the architecture. Finally, we define basic type and expression models that are used in MoT, MoD as well as ToD and thus simplify passing data from one model to another. First analysis shows that we can simplify building one generator up to a factor of 10x. This factor increases further when different target languages or target language styles are generated from one MoD as well.
KW - Design Productivity
KW - Hardware Generation
KW - Model Driven Architecture
KW - Model-of-Design (MoD)
KW - Model-of-Things (MoT)
UR - https://www.scopus.com/pages/publications/85006827166
U2 - 10.1109/VLSI-SoC.2016.7753576
DO - 10.1109/VLSI-SoC.2016.7753576
M3 - Conference contribution
AN - SCOPUS:85006827166
T3 - 2016 IFIP/IEEE International Conference on Very Large Scale Integration, VLSI-SoC 2016
BT - 2016 IFIP/IEEE International Conference on Very Large Scale Integration, VLSI-SoC 2016
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 24th Annual IFIP/IEEE International Conference on Very Large Scale Integration, VLSI-SoC 2016
Y2 - 26 September 2016 through 28 September 2016
ER -