Chuei-Tang Wang, Chia-Chia Lin, Chih-Hsin Lu, Wei-Ting Chen, C. Tsai, Douglas C. H. Yu
{"title":"SoIC_H Technology for Heterogenous System Integration","authors":"Chuei-Tang Wang, Chia-Chia Lin, Chih-Hsin Lu, Wei-Ting Chen, C. Tsai, Douglas C. H. Yu","doi":"10.1109/vlsitechnologyandcir46769.2022.9830182","DOIUrl":null,"url":null,"abstract":"An SoIC_H technology for 2.5D heterogeneous system integration is proposed. SoIC bond, replacing µbump, is used to provide low parasitic and high density of interconnects. Through system technology co-optimization (STCO), the proposed architecture provides over 60% power reduction for die-to-die I/O link and 81%, 14%, and 94% reductions in energy, latency, and area, respectively, for on-chip fanout design. For memory cubes, it introduces 61% latency and 49% energy reductions for 4-Hi SRAM cache and 30% bandwidth and 28% energy efficiency improvements for 12-Hi HBM.","PeriodicalId":332454,"journal":{"name":"2022 IEEE Symposium on VLSI Technology and Circuits (VLSI Technology and Circuits)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2022-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 IEEE Symposium on VLSI Technology and Circuits (VLSI Technology and Circuits)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/vlsitechnologyandcir46769.2022.9830182","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2
Abstract
An SoIC_H technology for 2.5D heterogeneous system integration is proposed. SoIC bond, replacing µbump, is used to provide low parasitic and high density of interconnects. Through system technology co-optimization (STCO), the proposed architecture provides over 60% power reduction for die-to-die I/O link and 81%, 14%, and 94% reductions in energy, latency, and area, respectively, for on-chip fanout design. For memory cubes, it introduces 61% latency and 49% energy reductions for 4-Hi SRAM cache and 30% bandwidth and 28% energy efficiency improvements for 12-Hi HBM.