A methodology for thermal simulation of interconnects enabled by model reduction with material property variation

• Published in: Journal of computational science 2022-05, Vol.61, p.101665, Article 101665
• Main Authors: Jia, Wangkun, Cheng, Ming-C.
• Format: Article
• Language: English
• Subjects: Data-driven; Interconnect thermal simulation; Method of snapshots; Model order reduction; Proper orthogonal decomposition
• ISSN: 1877-7503; 1877-7511
• DOI: 10.1016/j.jocs.2022.101665

A thermal simulation methodology is developed for interconnects enabled by a data-driven learning algorithm accounting for variations of material properties, heat sources and boundary conditions (BCs). The methodology is based on the concepts of model order reduction and domain decomposition to construct a multi-block approach. A generic block model is built to represent a group of interconnect blocks that are used to wire standard cells in the integrated circuits (ICs). The blocks in this group possess identical geometry with various metal/via routings. The data-driven model reduction method is thus applied to learn material property variations induced by different metal/via routings in the blocks, in addition to the variations of heat sources and BCs. The approach is investigated in two very different settings. It is first applied to thermal simulation of a single interconnect block with similar BCs to those in the training of the generic block. It is then implemented in multi-block thermal simulation of a FinFET IC, where the interconnect structure is partitioned into several blocks each modeled by the generic block model. Accuracy of the generic block model is examined in terms of the metal/via routings, BCs and thermal discontinuities at the block interfaces. •A data-driven POD approach is developed to capture material property variation (MPV).•Generic blocks are proposed to develop multi-block POD thermal models for interconnects.•Data quality is the key to construct accurate and robust MPV-POD thermal models.•MPV-POD model is able to offer a good prediction even beyond the training settings.•Model accuracy suffers from interface discontinuities that can be removed with more POD modes.