Zidong Du, K. Palem, L. Avinash, O. Temam, Yunji Chen, Chengyong Wu
{"title":"Leveraging the error resilience of machine-learning applications for designing highly energy efficient accelerators","authors":"Zidong Du, K. Palem, L. Avinash, O. Temam, Yunji Chen, Chengyong Wu","doi":"10.1109/ASPDAC.2014.6742890","DOIUrl":"https://doi.org/10.1109/ASPDAC.2014.6742890","url":null,"abstract":"In recent years, inexact computing has been increasingly regarded as one of the most promising approaches for reducing energy consumption in many applications that can tolerate a degree of inaccuracy. Driven by the principle of trading tolerable amounts of application accuracy in return for significant resource savings - the energy consumed, the (critical path) delay and the (silicon) area being the resources - this approach has been limited to certain application domains. In this paper, we propose to expand the application scope, error tolerance as well as the energy savings of inexact computing systems through neural network architectures. Such neural networks are fast emerging as popular candidate accelerators for future heterogeneous multi-core platforms, and have flexible error tolerance limits owing to their ability to be trained. Our results based on simulated 65nm technology designs demonstrate that the proposed inexact neural network accelerator could achieve 43.91%-62.49% savings in energy consumption (with corresponding delay and area savings being 18.79% and 31.44% respectively) when compared to existing baseline neural network implementation, at the cost of an accuracy loss (quantified as the Mean Square Error (MSE) which increases from 0.14 to 0.20 on average).","PeriodicalId":234635,"journal":{"name":"2014 19th Asia and South Pacific Design Automation Conference (ASP-DAC)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125690846","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
S. Ikeda, T. Kamimura, Sang-yeop Lee, Hiroyuki Ito, N. Ishihara, K. Masu
{"title":"A 950μW 5.5-GHz low voltage PLL with digitally-calibrated ILFD and linearized varactor","authors":"S. Ikeda, T. Kamimura, Sang-yeop Lee, Hiroyuki Ito, N. Ishihara, K. Masu","doi":"10.1109/ASPDAC.2014.6742855","DOIUrl":"https://doi.org/10.1109/ASPDAC.2014.6742855","url":null,"abstract":"This paper proposes an ultra-low-power 5.5-GHz PLL which employs a divide-by-4 injection-locked frequency divider (ILFD), which is calibrated by digital circuits, and linearity-compensated varactors for low supply-voltage operation. The proposed PLL was fabricated in 65nm CMOS. It shows a 1-MHz-offset phase noise of -106 dBc/Hz and the total power consumption of 950μW at 5.5 GHz.","PeriodicalId":234635,"journal":{"name":"2014 19th Asia and South Pacific Design Automation Conference (ASP-DAC)","volume":"68 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133135678","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Quantifying workload dependent reliability in embedded processors","authors":"V. Chandra","doi":"10.1109/ASPDAC.2014.6742936","DOIUrl":"https://doi.org/10.1109/ASPDAC.2014.6742936","url":null,"abstract":"With nearly three decades of continued CMOS scaling, the devices have now been pushed to their physical and reliability limits. Scaling to sub-20nm technology nodes changes the nature of reliability effects from abrupt functional problems to progressive degradation of the performance characteristics of devices and system components. The impact of unreliability results in time-dependent variability, directly translating into design uncertainty in manufactured chips. Further, application workloads can significantly affect the overall system reliability. In this work, we have analyzed aging effects on various design hierarchies of an embedded processor in 28nm running real-world applications. We have also quantified the dependencies of aging effects on switching-activity and power-state of workloads. Implementation results show that the processor timing degradation can vary from 2% to 11%, depending on the workload.","PeriodicalId":234635,"journal":{"name":"2014 19th Asia and South Pacific Design Automation Conference (ASP-DAC)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128131350","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"2-SAT based linear time optimum two-domain clock skew scheduling","authors":"Y. Kohira, A. Takahashi","doi":"10.1109/ASPDAC.2014.6742885","DOIUrl":"https://doi.org/10.1109/ASPDAC.2014.6742885","url":null,"abstract":"Multi-domain clock skew scheduling is an effective technique to improve the performance of sequential circuits by using practical clock distribution network. Although the upper bound of performance of a circuit increases as the number of clock domains increases in multi-domain clock skew scheduling, the improvement of the performance becomes smaller while the cost of clock distribution network increases much. In this paper, a linear time algorithm that finds an optimum two-domain clock skew schedule is proposed. Experimental results on ISCAS89 benchmark circuits and artificial data show that optimum circuits are efficiently obtained by our method in short time.","PeriodicalId":234635,"journal":{"name":"2014 19th Asia and South Pacific Design Automation Conference (ASP-DAC)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131346956","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Efficient feasibility analysis of DAG scheduling with real-time constraints in the presence of faults","authors":"Xiaotong Cui, Jun Zhang, Kaijie Wu, E. Sha","doi":"10.1109/ASPDAC.2014.6742878","DOIUrl":"https://doi.org/10.1109/ASPDAC.2014.6742878","url":null,"abstract":"Tasks in hard real-time systems are required to meet deadlines in the presence of faults. We conclude that a sufficient condition of a task set experiencing its worst-case finish time (WCFT) is that its critical task (CT) incurs all faults. An algorithm is presented to identify the CT and the WCFT in O(N2) with N being the task number. A common practice that bet the WCFT using the task with the longest re-execution time could under estimate by up-to 35%!","PeriodicalId":234635,"journal":{"name":"2014 19th Asia and South Pacific Design Automation Conference (ASP-DAC)","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127180400","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Spiking brain models: Computation, memory and communication constraints for custom hardware implementation","authors":"A. Lansner, A. Hemani, Nasim Farahini","doi":"10.1109/ASPDAC.2014.6742950","DOIUrl":"https://doi.org/10.1109/ASPDAC.2014.6742950","url":null,"abstract":"We estimate the computational capacity required to simulate in real time the neural information processing in the human brain. We show that the computational demands of a detailed implementation are beyond reach of current technology, but that some biologically plausible reductions of problem complexity can give performance gains between two and six orders of magnitude, which put implementations within reach of tomorrow's technology.","PeriodicalId":234635,"journal":{"name":"2014 19th Asia and South Pacific Design Automation Conference (ASP-DAC)","volume":"04 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127184029","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Routability-driven bump assignment for chip-package co-design","authors":"Meng-Ling Chen, Tu-Hsiung Tsai, Hung-Ming Chen, Shi-Hao Chen","doi":"10.1109/ASPDAC.2014.6742944","DOIUrl":"https://doi.org/10.1109/ASPDAC.2014.6742944","url":null,"abstract":"In current chip and package designs, it is a bottleneck to simultaneously optimize both pin assignment and pin routing for different design domains (chip, package, and board). Usually the whole process costs a huge manual effort and multiple iterations thus reducing profit margin. Therefore, we propose a fast heuristic chip-package co-design algorithm in order to automatically obtain a bump assignment which introduces high routability both in RDL routing and substrate routing (100% in our real case). Experimental results show that the proposed method (inspired by board escape routing algorithms) automatically finishes bump assignment, RDL routing and substrate routing in a short time, while the traditional co-design flow requires weeks even months.","PeriodicalId":234635,"journal":{"name":"2014 19th Asia and South Pacific Design Automation Conference (ASP-DAC)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132218322","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Structural planning of 3D-IC interconnects by block alignment","authors":"J. Knechtel, Evangeline F. Y. Young, J. Lienig","doi":"10.1109/ASPDAC.2014.6742866","DOIUrl":"https://doi.org/10.1109/ASPDAC.2014.6742866","url":null,"abstract":"Three-dimensional integrated circuits rely on optimized interconnect structures for blocks which are spread among one or multiple dies. We demonstrate how 2D and 3D block alignment can be efficiently utilized for structural planning of different interconnects. To realize this, we extend the corner block list and provide effective techniques for 3D layout generation, i.e., block placement and alignment. Our techniques are made available in an open-source, simulated-annealing-based tool. Besides block alignment, it accounts for key objectives in 3D design like fast thermal management and fixed-outline floorplanning. Experimental results on GSRC and IBM-HB+ circuits demonstrate the capabilities of our tool for both planning 3D-IC interconnects by block alignment and for 3D floorplanning in general.","PeriodicalId":234635,"journal":{"name":"2014 19th Asia and South Pacific Design Automation Conference (ASP-DAC)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132751908","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Chen-Yong Cher, K. Muller, R. Haring, D. Satterfield, T. Musta, Thomas Gooding, K. Davis, M. B. Dombrowa, G. Kopcsay, R. Senger, Yutaka Sugawara, K. Sugavanam
{"title":"Soft Error Resiliency Characterization on IBM BlueGene/Q Processor","authors":"Chen-Yong Cher, K. Muller, R. Haring, D. Satterfield, T. Musta, Thomas Gooding, K. Davis, M. B. Dombrowa, G. Kopcsay, R. Senger, Yutaka Sugawara, K. Sugavanam","doi":"10.1109/ASPDAC.2014.6742920","DOIUrl":"https://doi.org/10.1109/ASPDAC.2014.6742920","url":null,"abstract":"Soft Error Resiliency (SER) is a major concern for Petascale high performance computing (HPC) systems. In designing Blue Gene/Q (BG/Q) [8], many mechanisms were deployed to target SER including extensive use of Silicon-On-Insulator (SOI), radiation-hardened latches [7,13], detection and correction in on-chip arrays, and very low radiation packaging materials. On the other hand, it is well known that application behavior has major impacts on the masking (or “derating” factor) in system SER calculations. The principal goal of this project is to understand the interaction between BG/Q hardware and high-performance applications when it comes to SER by performing and evaluating a chip irradiation experiment.","PeriodicalId":234635,"journal":{"name":"2014 19th Asia and South Pacific Design Automation Conference (ASP-DAC)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128311969","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"The data center as a grid load stabilizer","authors":"Hao Chen, M. Caramanis, A. Coskun","doi":"10.1109/ASPDAC.2014.6742874","DOIUrl":"https://doi.org/10.1109/ASPDAC.2014.6742874","url":null,"abstract":"To accommodate the increasing presence of volatile and intermittent renewable energy sources in power generation, independent system operators (ISO) offer opportunities for demand side regulation service (RS) so as to stabilize the grid load. These power market features allow the demand side to earn monetary credits by modulating its power consumption dynamically following an RS signal broadcast by ISO. This paper studies the capacities and benefits of a major potential demand side, the data center, to provide RS. We propose a dynamic control policy that modulates the data center power consumption in response to ISO requests by leveraging server power capping techniques and various server power states. Results demonstrate that using our policy, data centers can provide fast reserves in quantities that are substantial proportions (around 50%) of their average energy consumption, with no major deterioration in quality of service (QoS). By doing so, data centers decrease their energy costs around 50%, while providing the ISOs and the society in general with cost effective demand side reserves that render massive renewable generation adoption affordable.","PeriodicalId":234635,"journal":{"name":"2014 19th Asia and South Pacific Design Automation Conference (ASP-DAC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128971856","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}