{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,5]],"date-time":"2026-06-05T15:34:07Z","timestamp":1780673647447,"version":"3.54.1"},"reference-count":39,"publisher":"Walter de Gruyter GmbH","issue":"1-2","license":[{"start":{"date-parts":[[2023,4,21]],"date-time":"2023-04-21T00:00:00Z","timestamp":1682035200000},"content-version":"unspecified","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2023,5,25]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Computation-in-Memory accelerators based on resistive switching devices represent a promising approach to realize future information processing systems. These architectures promise orders of magnitudes lower energy consumption for certain tasks, while also achieving higher throughputs than other special purpose hardware such as GPUs, due to their analog computation nature. Due to device variability issues, however, a single resistive switching cell usually does not achieve the resolution required for the considered applications. To overcome this challenge, many of the proposed architectures use an approach called bit slicing, where generally multiple low-resolution components are combined to realize higher resolution blocks. In this paper, we will present an analog accelerator architecture on the circuit level, which can be used to perform Vector-Matrix-Multiplications or Matrix-Matrix-Multiplications. The architecture consists of the 1T1R crossbar array, the optimized select circuitry and an ADC. The components are designed to handle the variability of the resistive switching cells, which is verified through our verified and physical compact model. We then use this architecture to compare different bit slicing approaches and discuss their tradeoffs.<\/jats:p>","DOI":"10.1515\/itit-2023-0018","type":"journal-article","created":{"date-parts":[[2023,4,21]],"date-time":"2023-04-21T21:38:01Z","timestamp":1682113081000},"page":"3-12","source":"Crossref","is-referenced-by-count":3,"title":["Bit slicing approaches for variability aware ReRAM CIM macros"],"prefix":"10.1515","volume":"65","author":[{"given":"Christopher","family":"Bengel","sequence":"first","affiliation":[{"name":"RWTH Aachen, Institute of Electronic Materials II , D-52074 Aachen , Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Leon","family":"Dixius","sequence":"additional","affiliation":[{"name":"RWTH Aachen, Institute of Electronic Materials II , D-52074 Aachen , Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rainer","family":"Waser","sequence":"additional","affiliation":[{"name":"RWTH Aachen, Institute of Electronic Materials II , D-52074 Aachen , Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dirk J.","family":"Wouters","sequence":"additional","affiliation":[{"name":"RWTH Aachen, Institute of Electronic Materials II , D-52074 Aachen , Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Stephan","family":"Menzel","sequence":"additional","affiliation":[{"name":"Forschungszentrum J\u00fclich, Peter Gr\u00fcnberg Institut 7 , D-52428 J\u00fclich , Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"374","published-online":{"date-parts":[[2023,4,21]]},"reference":[{"key":"2024082717154692153_j_itit-2023-0018_ref_001","doi-asserted-by":"crossref","unstructured":"Y. LeCun, Y. Bengio, and G. Hinton, \u201cDeep learning,\u201d Nature, vol.\u00a0521, pp.\u00a0436\u2013444, 2015. https:\/\/doi.org\/10.1038\/nature14539.","DOI":"10.1038\/nature14539"},{"key":"2024082717154692153_j_itit-2023-0018_ref_002","doi-asserted-by":"crossref","unstructured":"V. Sze, Y. Chen, J. Emer, A. Suleiman, and Z. Zhang, \u201cHardware for machine learning: challenges and opportunities,\u201d in 2018 IEEE Custom Integrated Circuits Conference (CICC), 2018, pp.\u00a01\u20138.","DOI":"10.1109\/CICC.2018.8357072"},{"key":"2024082717154692153_j_itit-2023-0018_ref_003","doi-asserted-by":"crossref","unstructured":"N. P. Jouppi, C. Young, N. Patil, et al.., \u201cIn-datacenter performance analysis of a tensor processing unit,\u201d SIGARCH Comput. Archit. News, vol.\u00a045, pp.\u00a01\u201312, 2017. https:\/\/doi.org\/10.1145\/3140659.3080246.","DOI":"10.1145\/3140659.3080246"},{"key":"2024082717154692153_j_itit-2023-0018_ref_004","doi-asserted-by":"crossref","unstructured":"Y. Jiao, L. Han, and X. Long, \u201cHanguang 800 npu \u2013 the ultimate ai inference solution for data centers,\u201d in 2020 IEEE Hot Chips 32 Symposium (HCS), 2020, pp.\u00a01\u201329.","DOI":"10.1109\/HCS49909.2020.9220619"},{"key":"2024082717154692153_j_itit-2023-0018_ref_005","unstructured":"N. C. Thompson, K. H. Greenewald, K. Lee, and G. F. Manso, \u201cThe computational limits of deep learning,\u201d arXiv:2007.05558v2, vol.\u00a0abs\/2007.05558, 2020."},{"key":"2024082717154692153_j_itit-2023-0018_ref_006","doi-asserted-by":"crossref","unstructured":"R. Dittmann, S. Menzel, and R. Waser, \u201cNanoionic memristive phenomena in metal oxides: the valence change mechanism,\u201d Adv. Phys., vol.\u00a070, no.\u00a02, pp.\u00a0155\u2013349, 2022. https:\/\/doi.org\/10.1080\/00018732.2022.2084006.","DOI":"10.1080\/00018732.2022.2084006"},{"key":"2024082717154692153_j_itit-2023-0018_ref_007","doi-asserted-by":"crossref","unstructured":"P. Chi, S. Li, C. Xu, et al.., \u201cPRIME: a novel processing-in-memory architecture for neural network computation in ReRAM-based main memory,\u201d in 2016 ACM\/IEEE 43rd Annual International Symposium on Computer Architecture (ISCA), 2016, pp.\u00a027\u201339.","DOI":"10.1109\/ISCA.2016.13"},{"key":"2024082717154692153_j_itit-2023-0018_ref_008","doi-asserted-by":"crossref","unstructured":"A. Shafiee, A. Nag, N. Muralimanohar, et al.., \u201cISAAC: a convolutional neural network accelerator with in-situ analog arithmetic in crossbars,\u201d in 2016 ACM\/IEEE 43rd Annual International Symposium on Computer Architecture (ISCA), 2016, pp.\u00a014\u201326.","DOI":"10.1109\/ISCA.2016.12"},{"key":"2024082717154692153_j_itit-2023-0018_ref_009","doi-asserted-by":"crossref","unstructured":"R. Waser, R. Dittmann, G. Staikov, and K. Szot, \u201cRedox-based resistive switching memories - nanoionic mechanisms, prospects, and challenges,\u201d Adv. Mater., vol.\u00a021, nos. 25\u201326, pp.\u00a02632\u20132663, 2009. https:\/\/doi.org\/10.1002\/adma.200900375.","DOI":"10.1002\/adma.200900375"},{"key":"2024082717154692153_j_itit-2023-0018_ref_010","doi-asserted-by":"crossref","unstructured":"D. B. Strukov, G. S. Snider, D. R. Stewart, and R. S. Williams, \u201cThe missing memristor found,\u201d Nature, vol.\u00a0453, no.\u00a07191, pp.\u00a080\u201383, 2008. https:\/\/doi.org\/10.1038\/nature06932.","DOI":"10.1038\/nature06932"},{"key":"2024082717154692153_j_itit-2023-0018_ref_011","doi-asserted-by":"crossref","unstructured":"M. A. Zidan, H. A. H. Fahmy, M. M. Hussain, and K. N. Salama, \u201cMemristor-based memory: the sneak paths problem and solutions,\u201d Microelectron. J., vol.\u00a044, no.\u00a02, pp.\u00a0176\u2013183, 2013. https:\/\/doi.org\/10.1016\/j.mejo.2012.10.001.","DOI":"10.1016\/j.mejo.2012.10.001"},{"key":"2024082717154692153_j_itit-2023-0018_ref_012","doi-asserted-by":"crossref","unstructured":"H. Kim, M. R. Mahmoodi, H. Nili, and D. B. Strukov, \u201c4K-memristor analog-grade passive crossbar circuit,\u201d Nat. Commun., vol.\u00a012, no.\u00a01, pp.\u00a05198\/1\u20135198\/11, 2021. https:\/\/doi.org\/10.1038\/s41467-021-25455-0.","DOI":"10.1038\/s41467-021-25455-0"},{"key":"2024082717154692153_j_itit-2023-0018_ref_013","doi-asserted-by":"crossref","unstructured":"C. Chou, Z. Lin, C. Lai, et al.., \u201cA 22\u202fnm 96KX144 RRAM macro with a self-tracking reference and a low ripple charge pump to achieve a configurable read window and a wide operating voltage range,\u201d in 2020 IEEE Symposium on VLSI Circuits, 2020, pp.\u00a01\u20132.","DOI":"10.1109\/VLSICircuits18222.2020.9163014"},{"key":"2024082717154692153_j_itit-2023-0018_ref_014","doi-asserted-by":"crossref","unstructured":"P. Jain, U. Arslan, M. Sekhar, et al.., \u201c13.2\u202fA 3.6 Mb 10.1 Mb\/mm2 embedded non-volatile ReRAM macro in 22\u202fnm Fin FET technology with adaptive forming\/set\/reset schemes yielding down to 0.5\u202fV with sensing time of 5\u202fns at 0.7 V,\u201d in 2019 IEEE International Solid- State Circuits Conference \u2013 (ISSCC), Intel, 2019, pp.\u00a0212\u2013214.","DOI":"10.1109\/ISSCC.2019.8662393"},{"key":"2024082717154692153_j_itit-2023-0018_ref_015","doi-asserted-by":"crossref","unstructured":"N. Kopperberg, S. Wiefels, K. Hofmann, et al.., \u201cEndurance of 2 mbit based beol integrated reram,\u201d IEEE Access, vol.\u00a010, pp.\u00a0122696\u2013122705, 2022. https:\/\/doi.org\/10.1109\/access.2022.3223657.","DOI":"10.1109\/ACCESS.2022.3223657"},{"key":"2024082717154692153_j_itit-2023-0018_ref_016","doi-asserted-by":"crossref","unstructured":"M. Horowitz, \u201cComputing\u2019s energy problem (and what we can do about it),\u201d in ISSCC 2014, pp.\u00a010\u201314, 2014.","DOI":"10.1109\/ISSCC.2014.6757323"},{"key":"2024082717154692153_j_itit-2023-0018_ref_017","doi-asserted-by":"crossref","unstructured":"A. Ankit, I. E. Hajj, S. R. Chalamalasetti, et al.., \u201cPANTHER: a programmable architecture for neural network training harnessing energy-efficient ReRAM,\u201d IEEE Trans. Comput., vol.\u00a069, pp.\u00a01128\u20131142, 2020. https:\/\/doi.org\/10.1109\/tc.2020.2998456.","DOI":"10.1109\/TC.2020.2998456"},{"key":"2024082717154692153_j_itit-2023-0018_ref_018","doi-asserted-by":"crossref","unstructured":"C. Li, J. Ignowski, X. Sheng, et al.., \u201cCMOS-integrated nanoscale memristive crossbars for CNN and optimization acceleration,\u201d in 2020 IEEE International Memory Workshop (IMW), 2020, pp.\u00a01\u20134.","DOI":"10.1109\/IMW48823.2020.9108112"},{"key":"2024082717154692153_j_itit-2023-0018_ref_019","doi-asserted-by":"crossref","unstructured":"J. J. Dongarra, J. D. Croz, S. Hammarling, and I. S. Duff, \u201cA set of level 3 basic linear algebra subprograms,\u201d ACM Trans. Math. Softw., vol.\u00a016, pp.\u00a01\u201317, 1990. https:\/\/doi.org\/10.1145\/77626.79170.","DOI":"10.1145\/77626.79170"},{"key":"2024082717154692153_j_itit-2023-0018_ref_020","doi-asserted-by":"crossref","unstructured":"B. Feinberg, U. K. R. Vengalam, N. Whitehair, S. Wang, and E. Ipek, \u201cEnabling scientific computing on memristive accelerators,\u201d in 2018 ACM\/IEEE 45th Annual International Symposium on Computer Architecture (ISCA), 2018, pp.\u00a0367\u2013382.","DOI":"10.1109\/ISCA.2018.00039"},{"key":"2024082717154692153_j_itit-2023-0018_ref_021","unstructured":"A. Ankit, I. E. Hajj, S. R. Chalamalasetti, et al.., \u201cA programmable ultra-efficient memristor-based accelerator for machine learning inference,\u201d in Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Systems, 2019, pp.\u00a0715\u2013731."},{"key":"2024082717154692153_j_itit-2023-0018_ref_022","doi-asserted-by":"crossref","unstructured":"W. Wan, R. Kubendran, C. Schaefer, et al.., \u201cA compute-in-memory chip based on resistive random-access memory,\u201d Nature, vol.\u00a0608, no.\u00a07923, pp.\u00a0504\u2013512, 2022. https:\/\/doi.org\/10.1038\/s41586-022-04992-8.","DOI":"10.1038\/s41586-022-04992-8"},{"key":"2024082717154692153_j_itit-2023-0018_ref_023","unstructured":"JART, \u201cJuelich aachen resistive switching tools (JART),\u201d Tech. Rep., IWE 2, RWTH Aachen, 2019."},{"key":"2024082717154692153_j_itit-2023-0018_ref_024","doi-asserted-by":"crossref","unstructured":"C. Bengel, J. Mohr, S. Wiefels, et al.., \u201cReliability aspects of binary vector-matrix-multiplications using ReRAM devices,\u201d Neuromorph. Comput. Eng., vol.\u00a02, no.\u00a03, p.\u00a0034001, 2022. https:\/\/doi.org\/10.1088\/2634-4386\/ac6d04.","DOI":"10.1088\/2634-4386\/ac6d04"},{"key":"2024082717154692153_j_itit-2023-0018_ref_025","doi-asserted-by":"crossref","unstructured":"C. Bengel, A. Siemon, F. C\u00fcppers, et al.., \u201cVariability-aware modeling of filamentary oxide based bipolar resistive switching cells using SPICE level compact models,\u201d IEEE Trans. Circuits Syst. I: Regul. Pap., vol.\u00a067, no.\u00a012, pp.\u00a04618\u20134630, 2020. https:\/\/doi.org\/10.1109\/tcsi.2020.3018502.","DOI":"10.1109\/TCSI.2020.3018502"},{"key":"2024082717154692153_j_itit-2023-0018_ref_026","doi-asserted-by":"crossref","unstructured":"S. Wiefels, C. Bengel, N. Kopperberg, K. Zhang, R. Waser, and S. Menzel, \u201cHRS instability in oxide based bipolar resistive switching cells,\u201d IEEE Trans. Electron Devices, vol.\u00a067, no.\u00a010, pp.\u00a04208\u20134215, 2020. https:\/\/doi.org\/10.1109\/ted.2020.3018096.","DOI":"10.1109\/TED.2020.3018096"},{"key":"2024082717154692153_j_itit-2023-0018_ref_027","doi-asserted-by":"crossref","unstructured":"F. C\u00fcppers, S. Menzel, C. Bengel, et al.., \u201cExploiting the switching dynamics of HfO2-based ReRAM devices for reliable analog memristive behavior,\u201d APL Mater., vol.\u00a07, no.\u00a09, pp.\u00a091105\/1\u201391105\/9, 2019. https:\/\/doi.org\/10.1063\/1.5108654.","DOI":"10.1063\/1.5108654"},{"key":"2024082717154692153_j_itit-2023-0018_ref_028","doi-asserted-by":"crossref","unstructured":"C. Bengel, F. C\u00fcppers, M. Payvand, et al.., \u201cUtilizing the switching stochasticity of HfO2\/TiOx-based ReRAM devices and the concept of multiple devices for the classification of overlapping and noisy patterns,\u201d Front. Neurosci., vol.\u00a015, p.\u00a0621, 2021.","DOI":"10.3389\/fnins.2021.661856"},{"key":"2024082717154692153_j_itit-2023-0018_ref_029","doi-asserted-by":"crossref","unstructured":"C. Bengel, A. Siemon, V. Rana, and S. Menzel, \u201cImplementation of multinary lukasiewicz logic using memristive devices,\u201d in 2021 IEEE International Symposium on Circuits and Systems (ISCAS), Daegu, Korea, 22\u201328 May, 2021, IEEE, 2021.","DOI":"10.1109\/ISCAS51556.2021.9401367"},{"key":"2024082717154692153_j_itit-2023-0018_ref_030","unstructured":"C. L. Torre, \u201cPhysics-based compact modeling of valence-change-based resistive switching devices,\u201d Ph.D. thesis, RWTH Aachen, 2019."},{"key":"2024082717154692153_j_itit-2023-0018_ref_031","doi-asserted-by":"crossref","unstructured":"E. Perez, M. K. Mahadevaiah, E. P. Quesada, and C. Wenger, \u201cVariability and energy consumption tradeoffs in multilevel programming of RRAM arrays,\u201d IEEE Trans. Electron Devices, vol.\u00a068, pp.\u00a02693\u20132698, 2021. https:\/\/doi.org\/10.1109\/ted.2021.3072868.","DOI":"10.1109\/TED.2021.3072868"},{"key":"2024082717154692153_j_itit-2023-0018_ref_032","doi-asserted-by":"crossref","unstructured":"P. Yao, H. Wu, B. Gao, et al.., \u201cFully hardware-implemented memristor convolutional neural network,\u201d Nature, vol.\u00a0577, no.\u00a07792, pp.\u00a0641\u2013646, 2020. https:\/\/doi.org\/10.1038\/s41586-020-1942-4.","DOI":"10.1038\/s41586-020-1942-4"},{"key":"2024082717154692153_j_itit-2023-0018_ref_033","doi-asserted-by":"crossref","unstructured":"A. Singh, M. A. Lebdeh, A. Gebregiorgis, R. Bishnoi, R. V. Joshi, and S. Hamdioui, \u201cSRIF: scalable and reliable integrate and fire circuit ADC for memristor-based CIM architectures,\u201d IEEE Trans. Circuits Syst. I: Regul. Pap., vol.\u00a068, pp.\u00a01\u201314, 2021. https:\/\/doi.org\/10.1109\/tcsi.2021.3061214.","DOI":"10.1109\/TCSI.2021.3061214"},{"key":"2024082717154692153_j_itit-2023-0018_ref_034","doi-asserted-by":"crossref","unstructured":"J. M. Hung, C. X. Xue, H. Y. Kao, et al.., \u201cA four-megabit compute-in-memory macro with eight-bit precision based on CMOS and resistive random-access memory for AI edge devices,\u201d Nat. Electron., vol.\u00a04, no.\u00a012, p.\u00a0921+, 2021. https:\/\/doi.org\/10.1038\/s41928-021-00676-9.","DOI":"10.1038\/s41928-021-00676-9"},{"key":"2024082717154692153_j_itit-2023-0018_ref_035","doi-asserted-by":"crossref","unstructured":"M. Zahedi, M. Mayahinia, M. A. Lebdeh, S. Wong, and S. Hamdioui, \u201cEfficient organization of digital periphery to support integer datatype for memristor-based CIM,\u201d in 2020 IEEE Computer Society Annual Symposium on VLSI (ISVLSI), 2020, pp.\u00a0216\u2013221.","DOI":"10.1109\/ISVLSI49217.2020.00047"},{"key":"2024082717154692153_j_itit-2023-0018_ref_036","doi-asserted-by":"crossref","unstructured":"M. Mayahinia, A. Singh, C. Bengel, et al.., \u201cA novel voltage controlled oscillation based ADC design for computation-in-memory using emerging ReRAMs,\u201d J. Emerg. Technol. Comput. Syst., vol.\u00a018, no.\u00a02, pp.\u00a01\u201325, 2022. https:\/\/doi.org\/10.1145\/3451212.","DOI":"10.1145\/3451212"},{"key":"2024082717154692153_j_itit-2023-0018_ref_037","doi-asserted-by":"crossref","unstructured":"T. P. Xiao, B. Feinberg, C. H. Bennett, et al.., \u201cAn accurate, error-tolerant, and energy-efficient neural network inference engine based on sonos analog memory,\u201d IEEE Trans. Circuits Syst. I: Regul. Pap., vol.\u00a069, pp.\u00a01480\u20131493, 2022. https:\/\/doi.org\/10.1109\/tcsi.2021.3134313.","DOI":"10.1109\/TCSI.2021.3134313"},{"key":"2024082717154692153_j_itit-2023-0018_ref_038","unstructured":"S. Wiefels, \u201cReliability aspects in resistively switching valence change memory cells,\u201d Ph.D. thesis, RWTH Aachen, 2021."},{"key":"2024082717154692153_j_itit-2023-0018_ref_039","doi-asserted-by":"crossref","unstructured":"B. Q. Le, A. Grossi, E. Vianello, et al.., \u201cResistive ram with multiple bits per cell: array-level demonstration of 3 bits per cell,\u201d IEEE Trans. Electron Devices, vol.\u00a066, no.\u00a01, pp.\u00a0641\u2013646, 2019. https:\/\/doi.org\/10.1109\/ted.2018.2879788.","DOI":"10.1109\/TED.2018.2879788"}],"container-title":["it - Information Technology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.degruyter.com\/document\/doi\/10.1515\/itit-2023-0018\/xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.degruyter.com\/document\/doi\/10.1515\/itit-2023-0018\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,27]],"date-time":"2024-08-27T17:16:17Z","timestamp":1724778977000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.degruyter.com\/document\/doi\/10.1515\/itit-2023-0018\/html"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,4,21]]},"references-count":39,"journal-issue":{"issue":"1-2","published-online":{"date-parts":[[2023,4,21]]},"published-print":{"date-parts":[[2023,5,25]]}},"alternative-id":["10.1515\/itit-2023-0018"],"URL":"https:\/\/doi.org\/10.1515\/itit-2023-0018","relation":{},"ISSN":["1611-2776","2196-7032"],"issn-type":[{"value":"1611-2776","type":"print"},{"value":"2196-7032","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,4,21]]}}}