Pith. sign in

REVIEW 4 major objections 6 minor 68 references

Architectural Exploration of Application-Specific Resonant SRAM Compute-in-Memory (rCiM)

T0 review · 4 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

Pith's one-line read An automated tool maps combinational logic onto resonant SRAM compute-in-memory arrays and selects the most energy-efficient macro topology, reporting average savings of 80.9% over a single-macro baseline.

desk verdict A plausible new exploration flow over AIG transforms and SRAM topologies, undermined by an internally inconsistent headline energy-savings figure that the paper never reconciles. read the letter →

arxiv 2411.09546 v1 pith:LF7QCFXU submitted 2024-11-14 cs.AR cs.CYcs.ETcs.SYeess.SY

classification cs.ARcs.CYcs.ETcs.SYeess.SY
keywords compute-in-memorySRAMresonantenergyrecyclingarchitecturalexplorationlogicsynthesisAIG10Tbitcelloptimization
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that a design tool can automatically pick the most energy-efficient SRAM compute-in-memory (CiM) configuration for a given combinational logic circuit, and that splitting one large memory macro into several smaller macros that execute NAND, NOR, and NOT operations in parallel saves energy and latency. The tool synthesizes many logic-graph variants of the circuit, maps each onto up to twelve SRAM macro topologies, estimates power and latency from characterized per-operation data, and returns the lowest-energy option along with a resonant-inductor size. The paper reports substantial average energy reductions for multi-macro over single-macro implementations across nine benchmark circuits, with the abstract's headline figure being an 80.9% average reduction for the six-topology configuration. A sympathetic reader would care because this is a concrete step toward closing the design-automation gap that has slowed adoption of in-memory computing.

What carries the argument

The load-bearing pieces are a 10-transistor SRAM bitcell with a dedicated dual read port (so two operands can be accessed without read disturb), a resonant write driver that recycles bitline discharge energy through a series inductor, and an exploration loop that combines the ABC logic synthesizer's AIG transformations with a linear energy and latency model. The evaluation model characterizes power and latency from post-layout simulations for each operation type and macro size, then scales them linearly by the number of operations and logic levels, with the inductor size chosen to match the macro's bitline capacitance. This machinery turns the architectural question of how many macros of what size into a finite search over 6,912 implementation strategies.

What would settle it

Run a full post-layout simulation of one benchmark (for example, the adder) on a single-macro and a three-macro configuration at the same macro size and compare measured energy against the tool's estimate; if the three-macro configuration is not the lower-energy one, the linear-scaling model fails.

Watch

Extended reading notes

Core claim

The central claim is that a given combinational circuit's best implementation in the resonant compute-in-memory (rCiM) architecture can be found automatically rather than by manual architectural choice. The tool enumerates 64 and-inverter-graph (AIG) variants produced by ordered combinations of four logic-synthesis transformations, evaluates each against twelve SRAM topologies (macro sizes 4 KB to 192 KB, with one, three, or six macros), and selects the configuration with the lowest estimated energy along with the resonant-inductor value. The paper argues that splitting a large macro into three parallel macros, each dedicated to NAND2, NOR2, or NOT, cuts energy by about 39% on average relative to a single macro because the same number of operations finishes in fewer cycles, and that the six-macro configuration trades higher power for further latency reduction (66% lower than single-macro latency) at an energy cost 15% above the three-macro case. The abstract's headline figure is an 80.9% average energy reduction for the six-topology implementation over a single-macro baseline across cache sizes from 4 KB to 192 KB.

Load-bearing premise

The tool's energy and latency numbers assume that power per operation is constant and that total latency is simply the number of logic levels times a fixed cycle time, with no extra control or routing overhead when the number of macros grows.

Editorial extensions

If this is right

  • If the tool's energy estimates are accurate, circuit designers could hand a Verilog description to the tool and receive a memory-macro size, macro count, and inductor value without manual architectural exploration.
  • Multi-macro execution of NAND2/NOR2/NOT operations would let the same SRAM array serve as a programmable logic block, not just a store, cutting latency for combinational workloads.
  • Energy recycling through the resonant write driver would make the writeback cycle nearly free in energy, which is the operation that dominates CiM cost.
  • The tool's linear scaling model, if validated, would let the search extend to larger macro counts and memory sizes without re-characterizing every configuration.
  • Choosing among AIG variants with different levels and gate counts gives a new joint optimization knob: logic synthesis and memory topology can be co-optimized.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same search methodology could be applied to other CiM cell types (for example, 6T or 8T with different logic primitives) by swapping the characterization data, making the exploration flow a general template rather than a one-off.
  • The linear scaling of power with operation count ignores potential overheads from address decoding and control that grow with macro count, so the reported savings may be optimistic for very small or very large macro counts.
  • The tradeoff between three and six macros suggests a sweet spot that depends on the circuit's ratio of logic levels to operations; the tool could be used to classify circuits as latency-bound or energy-bound and tune synthesis accordingly.
  • If energy recycling is as effective as simulated, the rCiM approach could lower the barrier to embedding boolean computation in last-level caches, where write energy is currently a blocker.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. This paper presents an architectural exploration tool that maps combinational RTL onto a resonant SRAM compute-in-memory (rCiM) architecture. The rCiM macro uses a 10T dual-read-port bitcell to perform NAND2/NOR2/NOT operations, a single-ended sense amplifier, and a series-resonant inductor write driver for energy recycling. The tool generates 64 AIG variants per benchmark via ABC/YOSYS synthesis transformations, evaluates them across 12 SRAM topologies (single-, three-, and six-macro configurations at sizes from 4KB to 192KB), and selects an energy/latency-optimal configuration for nine EPFL benchmark circuits in TSMC 28nm. The abstract claims an average 80.9% energy saving for the six-topology implementation over a single-macro baseline; the body instead reports three-macro implementations at 39% lower energy than single-macro and six-macro at 15% higher energy than three-macro, while the conclusion reports 40.52% and Table I reports 89.12% for best-case versus worst-case. The paper also provides transient simulations and 5000-sample Monte-Carlo analyses of the NAND2/NOR2 sensing operation, plus comparisons of throughput and energy efficiency with prior CiM macros.

Significance. If the energy-saving claims were consistent and the underlying energy model were validated, this would be a useful design-space exploration tool for SRAM-based CiM, combining synthesis-level AIG transformations with macro-level characterization. The paper's strengths are the breadth of the exploration (6912 implementations across 12 topologies and nine benchmarks), the detailed Monte-Carlo variation data for the sensing path, and the integration of established open-source synthesis tools with a custom characterization flow. However, the core quantitative contribution is currently not reliable: the headline savings figure in the abstract is not supported by the body, the reported savings figures are mutually inconsistent, and the analytical energy model used to produce them is not documented or validated. The significance of the contribution is therefore contingent on a substantial revision that reconciles the numbers and makes the energy-accounting layer reproducible.

major comments (4)
  1. [Abstract, Section IV-B, Conclusion, Table I] The headline claim of an average 80.9% energy reduction with the six-topology implementation is not present in the body and is contradicted by the paper's own numbers. Section IV-B and Figure 9(c) state that three-macro implementations use 39% lower energy than single-macro implementations and that six-macro implementations use 15% higher energy than three-macro implementations, which implies roughly 30% savings for six-macro versus single-macro. The Conclusion reports 40.52% for three-macro versus single-macro at the same macro size, and the Table I caption reports 89.12% for best-case versus worst-case. These figures cannot all be correct, and because six-macro is reported as more energy-hungry than three-macro, the abstract's 'six-topology implementation' is inconsistent with an energy-minimizing tool selecting the best topology. Please reconcile all reported savings, define the exact comparison (same macro size, same synthesis transformations, same total memory footprint), and provide per-benchmark energy data.
  2. [Algorithm I (lines 10-13) and Section IV-B] The energy and latency evaluation is an undocumented analytical model. The text asserts that power is identical for single- and three-macro implementations because the total number of operations is constant, that six-macro power is double three-macro power because power per cycle doubles while the cycle count is unchanged, and implicitly that latency equals the number of AIG levels times the cycle time. No closed-form energy equations are given, no scaling rules for characterized per-operation power/latency values are specified, and the model is not validated against full-array SPICE or post-layout simulations. Since all savings in Figure 9 and Table I are computed through this evaluation step, the central energy results are not independently verifiable as reported.
  3. [Table I] The 89.12% average saving stated in the Table I caption is a best-case-versus-worst-case comparison that confounds three variables: macro count, macro size, and synthesis transformation. Every best-case row is a three-macro configuration with 16KB or 32KB SRAM, while every worst-case row is a single-macro 4KB implementation, and the synthesis recipes differ as well. This does not isolate the effect of the multi-macro topology. Please replace this with an apples-to-apples comparison that varies only the macro count while holding macro size and synthesis transformation fixed, or explicitly decompose the contributions of topology, size, and synthesis recipe.
  4. [Section II and Section IV-D] The resonant write driver and the rCiM macro are adopted from references [51], [52], and [58], but the manuscript does not include an independent comparison of the resonant write-back path against a conventional write driver in the same 28nm implementation. The per-operation energy figures in Section IV-D (65 fJ per NAND2 and 116 fJ per NOR2) and the energy-efficiency comparisons in Table II therefore do not isolate the contribution of the series-resonant energy recycling that is presented as a main contribution. A direct simulation or measurement of write energy with and without the resonant driver would establish the claimed benefit.
minor comments (6)
  1. [Throughout] The manuscript misspells 'von Neumann' as 'V on Neumann' in the Abstract, Section I, and Figure 1; please correct this across the text.
  2. [Abstract and Section IV-B] The terminology 'six-topology implementation' in the abstract is not used in the body, which refers to 'six-macro' topologies; please define precisely which of the 12 topologies are single-, three-, and six-macro and use consistent names throughout.
  3. [Table II] Table II is difficult to read because several cells contain multiple unlabeled values, and the normalization of throughput to an 8KB array and the process scaling via Dennard's law are not described in sufficient detail to reproduce the comparison.
  4. [Section IV-C] The Monte-Carlo analysis covers the NAND2/NOR2 sensing path but not the resonant write driver, pulse generator, or sense-amplifier offset; a sentence clarifying the scope of the variation analysis would prevent overgeneralization.
  5. [Algorithm I, line 9] The heuristic that the memory size must be at least four times the gate count (2 inputs plus 2 outputs per gate) is asserted without derivation; please justify it or provide a sensitivity analysis.
  6. [Section III-D] There are missing cross-references and typos: the text refers to 'Figur. 8' and 'Algorithm' without a number, and the acronym 'rCIM' appears in Algorithm I while 'rCiM' is used elsewhere.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the tool's energy and latency estimates come from the paper's own characterized simulations and post-layout analysis, not from fitting the reported savings, so the exploration claim is not circular; the internal inconsistencies in the 80.9% figure are correctness issues, not circularity.

full rationale

The paper's derivation chain consists of AIG synthesis via ABC/Yosys, enumeration of rCiM topologies, and an energy/latency evaluation step in Algorithm I. Lines 10-13 compute power, latency, and energy metrics from post-layout SRAM characterization and analytical estimation; Line 14 then selects the lowest-energy configuration from these precomputed metrics. No parameter is fitted to the reported energy savings, and no equation shown in the paper reduces to its own input by construction. The energy-recycling write driver is adopted from self-cited references [51], [52], and [58], but the paper also includes its own SPICE transient simulations, Monte-Carlo robustness analysis, and per-benchmark energy comparisons, so the design is not justified solely by self-citation. The abstract's 80.9% figure conflicts with the conclusion's 40.52%, Table I's 89.12% confounds best-versus-worst comparisons, and Section IV-B implies roughly 30% savings; these are internal numeric and methodological inconsistencies that undermine the headline quantitative claim, but they are not circularity. No self-definitional, fitted-input, or imported-uniqueness pattern was found, so the circularity score is 0.

Assumptions & free parameters 4 free parameters · 4 assumptions · 2 invented entities

The central claim rests on a small set of hand-chosen design parameters, on an unstated linear-scaling energy model, and on self-cited prior hardware work. The paper introduces circuit topologies rather than physically new entities, and those circuit topologies have simulation-only support.

free parameters (4)
  • NAND2 read-wordline pulse width = ~150 ps
    Chosen so that Rbl discharges below Vref for '00' but remains above Vref for '10/01'; central to NAND2 sensing and to per-operation latency and energy.
  • NOR2 read-wordline pulse width = ~350 ps
    Chosen so that Rbl fully discharges for '10/01'; distinguishes NOR2 from NAND2 and affects operation latency and energy.
  • Vref level for sense amplifier and resonant write driver = VDD/2
    Set at mid-rail to maximize sensing margin and energy recycling; affects correctness and the reported energy savings.
  • SRAM sizing heuristic factor = 4x gate count
    Algorithm I line 9 selects memory size as at least four times the number of gates, accounting for 2 inputs and 2 outputs per gate. This heuristic determines the candidate topology set and therefore the reported optimal energy.
assumptions (4)
  • domain assumption Series LC resonance recycling with Vref equal to VDD/2 recovers write-bitline energy with zero net current.
    Adopted from self-cited references [51]-[57] without re-derivation; the reported write-energy savings depend on this mechanism.
  • domain assumption The 10T dual-read-port bitcell and sense amplifier operate correctly at 1 GHz with the characterized pulse widths in TSMC 28nm.
    Supported only by SPICE transient and Monte Carlo simulations in this paper; there is no fabricated silicon measurement.
  • domain assumption Each AIG logic level maps to one clock cycle and all operations in a level can be executed in parallel within a macro, with write-back in a subsequent cycle.
    This is the basis of the latency model in Section III-D and the scaling arguments in Section IV-B.
  • domain assumption Power, latency, and energy of arbitrary benchmark circuits can be estimated by linearly scaling characterized per-operation values across macro count and size.
    Invoked in Algorithm I lines 10 to 13 and Section IV-B; no closed-form model or validation against full-array simulation is provided.
invented entities (2)
  • 10T dual-read-port SRAM bitcell
    purpose: Enables NAND2, NOR2, and NOT operations with decoupled operand access and reduced read-disturb risk.
    Only SPICE transient and Monte Carlo simulations are provided; no fabricated silicon measurement or external validation supports the claims.
  • Shared series resonant inductor write driver
    purpose: Recycles write-bitline energy into a VDD/2 node to reduce dynamic write power.
    Adapted from self-cited prior works [51]-[57] and validated here only through simulation; no independent silicon data appears in this paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Architectural Exploration of Application-Specific Resonant SRAM Compute-in-Memory (rCiM)." pith.science (2026). https://pith.science/paper/LF7QCFXU

@misc{pith2026241109546,
  author       = {Pith},
  title        = {Pith review of: Architectural Exploration of Application-Specific Resonant SRAM Compute-in-Memory (rCiM)},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LF7QCFXU}},
  note         = {Machine review of arXiv:2411.09546}
}
read the original abstract

While general-purpose computing follows Von Neumann's architecture, the data movement between memory and processor elements dictates the processor's performance. The evolving compute-in-memory (CiM) paradigm tackles this issue by facilitating simultaneous processing and storage within static random-access memory (SRAM) elements. Numerous design decisions taken at different levels of hierarchy affect the figure of merits (FoMs) of SRAM, such as power, performance, area, and yield. The absence of a rapid assessment mechanism for the impact of changes at different hierarchy levels on global FoMs poses a challenge to accurately evaluating innovative SRAM designs. This paper presents an automation tool designed to optimize the energy and latency of SRAM designs incorporating diverse implementation strategies for executing logic operations within the SRAM. The tool structure allows easy comparison across different array topologies and various design strategies to result in energy-efficient implementations. Our study involves a comprehensive comparison of over 6900+ distinct design implementation strategies for EPFL combinational benchmark circuits on the energy-recycling resonant compute-in-memory (rCiM) architecture designed using TSMC 28 nm technology. When provided with a combinational circuit, the tool aims to generate an energy-efficient implementation strategy tailored to the specified input memory and latency constraints. The tool reduces 80.9% of energy consumption on average across all benchmarks while using the six-topology implementation compared to baseline implementation of single-macro topology by considering the parallel processing capability of rCiM cache size ranging from 4KB to 192KB.

Figures

Figures reproduced from arXiv: 2411.09546 by the authors.

Figure 1
Figure 1. (a) Conventional Von Neumann architecture, where an operation [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. (a) Conventional SRAM write driver exhibits high dynamic power [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. (a) The schematic of the proposed 10T SRAM cell and (b) the [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: 10T-SRAM bitcell along with resonant write driver implementing [PITH_FULL_IMAGE:figures/full_fig_p003_4.png]
Figure 5
Figure 5. Figure 5: shows the transient simulation of performing a single NAND2 operation for cases “10/01”. The read bit￾lines, RblA & RblB, are connected to one end of a single￾ended sense amplifier (SA) through the column mux switches (col muxA & col muxB ) as shown in [PITH_FULL_IMAG…
Figure 6
Figure 6. Figure 6: Comparison of memory topology considerations for rCiM architecture, [PITH_FULL_IMAGE:figures/full_fig_p005_6.png]
Figure 7
Figure 7. Figure 7: The AIG graph generated using different synthesis transformations results in AIGs with different levels and different numbers of gates at each level [PITH_FULL_IMAGE:figures/full_fig_p006_7.png]
Figure 8
Figure 8. Figure 8: The proposed methodology flow chart shows different operations in [PITH_FULL_IMAGE:figures/full_fig_p007_8.png]
Figure 9
Figure 9. Figure 9: After mapping each benchmark circuit to different SRAM architectures, we computed the power, latency, and energy; (a) power consumption remained [PITH_FULL_IMAGE:figures/full_fig_p008_9.png]
Figure 10
Figure 10. Figure 10: Monte-Carlo simulations considering 5000 samples of the [PITH_FULL_IMAGE:figures/full_fig_p010_10.png]
Figure 11
Figure 11. Figure 11: Monte-Carlo simulations with variations in the temperature and [PITH_FULL_IMAGE:figures/full_fig_p011_11.png]
Figure 12
Figure 12. Figure 12: Process variation analysis of the readout circuit considering all the cases for NAND2 and NOR2 operations show successful computational results [PITH_FULL_IMAGE:figures/full_fig_p012_12.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

68 extracted references · 63 canonical work pages

  1. [51]

    Resonant Energy Recycling SRAM Architecture,

    R. Islam, B. Saha, and I. Bezzam, “Resonant Energy Recycling SRAM Architecture,” IEEE Transactions on Circuits and Systems II: Express Briefs, vol. 68, no. 4, pp. 1383–1387, 2021

  2. [58]

    Resonant Compute- In-Memory (rCIM) 10T SRAM Macro for Boolean Logic,

    D. Challagundla, I. Bezzam, B. Saha, and R. Islam, “Resonant Compute- In-Memory (rCIM) 10T SRAM Macro for Boolean Logic,” in IEEE 41st International Conference on Computer Design (ICCD) , 2023, pp. 110– 117

  3. [52]

    A Low V oltage SRAM Using Resonant Supply Boosting,

    R. V . Joshi, M. M. Ziegler, and H. Wetter, “A Low V oltage SRAM Using Resonant Supply Boosting,” IEEE Journal of Solid-State Circuits, vol. 52, no. 3, pp. 634–644, 2017

  4. [1]

    A 65 nm 1.4-6.7 TOPS/W Adaptive-SNR Sparsity-Aware CIM Core with Load Balancing Support for DL work- loads,

    M. Ali, I. Chakraborty, S. Choudhary, M. Chang, D. E. Kim, A. Ray- chowdhury, and K. Roy, “A 65 nm 1.4-6.7 TOPS/W Adaptive-SNR Sparsity-Aware CIM Core with Load Balancing Support for DL work- loads,” in IEEE Custom Integrated Circuits Conference (CICC) , 2023, pp. 1–2

  5. [2]

    EASI-CiM: Event-driven Asynchronous Stream-based Im- age classifier with Compute-in-Memory kernels,

    R. Sreekumar, M. Park, M. N. Sakib, B. S. Reniwal, K. Lee, and M. R. Stan, “EASI-CiM: Event-driven Asynchronous Stream-based Im- age classifier with Compute-in-Memory kernels,” in 25th International Symposium on Quality Electronic Design (ISQED) , 2024, pp. 1–8

  6. [3]

    A Charge Domain SRAM Compute-in-Memory Macro With C-2C Ladder-Based 8-Bit MAC Unit in 22-nm FinFET Process for Edge Inference,

    H. Wang, R. Liu, R. Dorrance, D. Dasalukunte, D. Lake, and B. Carlton, “A Charge Domain SRAM Compute-in-Memory Macro With C-2C Ladder-Based 8-Bit MAC Unit in 22-nm FinFET Process for Edge Inference,” IEEE Journal of Solid-State Circuits , vol. 58, no. 4, pp. 1037–1050, 2023

  7. [4]

    A Local Computing Cell and 6T SRAM-Based Computing-in- Memory Macro With 8-b MAC Operation for Edge AI Chips,

    X. Si, Y .-N. Tu, W.-H. Huang, J.-W. Su, P.-J. Lu, J.-H. Wang, T.-W. Liu, S.-Y . Wu, R. Liu, Y .-C. Chou, Y .-L. Chung, W. Shih, C.-C. Lo, R.-S. Liu, C.-C. Hsieh, K.-T. Tang, N.-C. Lien, W.-C. Shih, Y . He, Q. Li, and M.-F. Chang, “A Local Computing Cell and 6T SRAM-Based Computing-in- Memory Macro With 8-b MAC Operation for Edge AI Chips,” IEEE Journal o...

  8. [5]

    A Variation-Tolerant In-Memory Machine Learning Classifier via On-Chip Training,

    S. K. Gonugondla, M. Kang, and N. R. Shanbhag, “A Variation-Tolerant In-Memory Machine Learning Classifier via On-Chip Training,” IEEE Journal of Solid-State Circuits , vol. 53, no. 11, pp. 3163–3173, 2018

Show all 68 references
  1. [6]

    IMAC: In-memory multi-bit multiplication and ACcumulation in 6T SRAM array,

    M. Ali, A. Jaiswal, S. Kodge, A. Agrawal, I. Chakraborty, and K. Roy, “IMAC: In-memory multi-bit multiplication and ACcumulation in 6T SRAM array,” IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 67, no. 8, pp. 2521–2531, 2020

  2. [7]

    A 2941-TOPS/W Charge-Domain 10T SRAM Compute-in-Memory for Ternary Neural Network,

    S. Cheon, K. Lee, and J. Park, “A 2941-TOPS/W Charge-Domain 10T SRAM Compute-in-Memory for Ternary Neural Network,” IEEE Transactions on Circuits and Systems I: Regular Papers , vol. 70, no. 5, pp. 2085–2097, 2023

  3. [8]

    A 28 nm 16 Kb Bit-Scalable Charge-Domain Transpose 6T SRAM In- Memory Computing Macro,

    J. Song, X. Tang, X. Qiao, Y . Wang, R. Wang, and R. Huang, “A 28 nm 16 Kb Bit-Scalable Charge-Domain Transpose 6T SRAM In- Memory Computing Macro,” IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 70, no. 5, pp. 1835–1845, 2023

  4. [9]

    Conv-RAM: An energy-efficient SRAM with embedded convolution computation for low-power CNN- based machine learning applications,

    A. Biswas and A. P. Chandrakasan, “Conv-RAM: An energy-efficient SRAM with embedded convolution computation for low-power CNN- based machine learning applications,” in IEEE International Solid - State Circuits Conference - (ISSCC) , 2018, pp. 488–490

  5. [10]

    A Resonant Time- Domain Compute-in-Memory (rTD-CiM) ADC-Less Architecture for MAC Operations,

    D. Challagundla, I. Bezzam, and R. Islam, “A Resonant Time- Domain Compute-in-Memory (rTD-CiM) ADC-Less Architecture for MAC Operations,” in Proceedings of the Great Lakes Symposium on VLSI , ser. GLSVLSI ’24. New York, NY , USA: Association for Computing Machinery, 2024, p. 2...

  6. [11]

    Design and Analysis of Multibit Multiply and Accumulate (MAC) unit: An Analog In-Memory Computing Approach,

    S. Ananthanarayanan, B. S. Reniwal, and A. Upadhyay, “Design and Analysis of Multibit Multiply and Accumulate (MAC) unit: An Analog In-Memory Computing Approach,” in 36th International Conference on VLSI Design and 22nd International Conference on Embedded Systems (VLSID), 202...

  7. [12]

    CiMComp: An Energy Effi- cient Compute-in-Memory Based Comparator for Convolutional Neural Networks,

    K. S, B. J. Kailath, and B. S. Reniwal, “CiMComp: An Energy Effi- cient Compute-in-Memory Based Comparator for Convolutional Neural Networks,” in Design, Automation and Test in Europe Conference and Exhibition (DATE), 2024, pp. 1–2

  8. [13]

    A 28-nm Floating-Point Computing-in-Memory Processor Using Intensive-CIM Sparse-Digital Architecture,

    S. Yan, J. Yue, C. He, Z. Wang, Z. Cong, Y . He, M. Zhou, W. Sun, X. Li, C. Dou, F. Zhang, H. Yang, Y . Liu, and M. Liu, “A 28-nm Floating-Point Computing-in-Memory Processor Using Intensive-CIM Sparse-Digital Architecture,” IEEE Journal of Solid-State Circuits , pp. 1–14, 2024

  9. [14]

    A 129.83 TOPS/W Area Efficient Digital SOT/STT MRAM-Based Computing-In-Memory for Advanced Edge AI Chips,

    L. Lu, A. Mani, and A. T. Do, “A 129.83 TOPS/W Area Efficient Digital SOT/STT MRAM-Based Computing-In-Memory for Advanced Edge AI Chips,” in IEEE International Symposium on Circuits and Systems (ISCAS), 2023, pp. 1–5. IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS 13

  10. [15]

    An 8b-Precision 6T SRAM Computing-in- Memory Macro Using Time-Domain Incremental Accumulation for AI Edge Chips,

    P.-C. Wu, J.-W. Su, Y .-L. Chung, L.-Y . Hong, J.-S. Ren, F.-C. Chang, Y . Wu, H.-Y . Chen, C.-H. Lin, H.-M. Hsiao, S.-H. Li, S.-S. Sheu, S.- C. Chang, W.-C. Lo, C.-I. Wu, C.-C. Lo, R.-S. Liu, C.-C. Hsieh, K.-T. Tang, and M.-F. Chang, “An 8b-Precision 6T SRAM Computing-in- Mem...

  11. [16]

    SUN: Dynamic Hybrid-Precision SRAM-Based CIM Accelerator With High Macro Utilization Using Structured Pruning Mixed-Precision Networks,

    Y .-W. Chen, R.-H. Wang, Y .-H. Cheng, C.-C. Lu, M.-F. Chang, and K.-T. Tang, “SUN: Dynamic Hybrid-Precision SRAM-Based CIM Accelerator With High Macro Utilization Using Structured Pruning Mixed-Precision Networks,” IEEE Transactions on Computer-Aided Design of Integrated Circ...

  12. [17]

    A Hierarchically Reconfigurable SRAM-Based Compute-in-Memory Macro for Edge Computing,

    R. Wang and X. Guo, “A Hierarchically Reconfigurable SRAM-Based Compute-in-Memory Macro for Edge Computing,” in IEEE 5th In- ternational Conference on Artificial Intelligence Circuits and Systems (AICAS), 2023, pp. 1–5

  13. [18]

    Impact of Aging and Process Vari- ability on SRAM-Based In-Memory Computing Architectures,

    J. B. Shaik, X. Guo, and S. Singhal, “Impact of Aging and Process Vari- ability on SRAM-Based In-Memory Computing Architectures,” IEEE Transactions on Circuits and Systems I: Regular Papers , vol. 71, no. 6, pp. 2696–2708, 2024

  14. [19]

    An Area-Efficient In-Memory Implementation Method of Arbitrary Boolean Function Based on SRAM Array,

    S. Zhang, X. Cui, F. Wei, and X. Cui, “An Area-Efficient In-Memory Implementation Method of Arbitrary Boolean Function Based on SRAM Array,” IEEE Transactions on Computers , vol. 72, no. 12, pp. 3416– 3430, 2023

  15. [20]

    PIC-RAM: Process- Invariant Capacitive Multiplier Based Analog In Memory Computing in 6T SRAM,

    K. Prasad, A. Biswas, A. Kabra, and J. Mekie, “PIC-RAM: Process- Invariant Capacitive Multiplier Based Analog In Memory Computing in 6T SRAM,” in Design, Automation and Test in Europe Conference and Exhibition (DATE), 2023, pp. 1–6

  16. [21]

    En- abling Energy-Efficient In-Memory Computing With Robust Assist- Based Reconfigurable Sense Amplifier in SRAM Array,

    K. Soundrapandiyan, S. K. Vishvakarma, and B. S. Reniwal, “En- abling Energy-Efficient In-Memory Computing With Robust Assist- Based Reconfigurable Sense Amplifier in SRAM Array,” IEEE Journal on Emerging and Selected Topics in Circuits and Systems, vol. 13, no. 1, pp. 445–455, 2023

  17. [22]

    14.2 A Compute SRAM with Bit-Serial Integer/Floating- Point Operations for Programmable In-Memory Vector Acceleration,

    J. Wang, X. Wang, C. Eckert, A. Subramaniyan, R. Das, D. Blaauw, and D. Sylvester, “14.2 A Compute SRAM with Bit-Serial Integer/Floating- Point Operations for Programmable In-Memory Vector Acceleration,” in IEEE International Solid- State Circuits Conference - (ISSCC) , 2019, ...

  18. [23]

    Analysis and Optimization Strategies Toward Reliable and High-Speed 6T Compute SRAM,

    J. Chen, W. Zhao, Y . Wang, and Y . Ha, “Analysis and Optimization Strategies Toward Reliable and High-Speed 6T Compute SRAM,” IEEE Transactions on Circuits and Systems I: Regular Papers , vol. 68, no. 4, pp. 1520–1531, 2021

  19. [24]

    A Reliable 8T SRAM for High-Speed Searching and Logic-in-Memory Operations,

    J. Chen, W. Zhao, Y . Wang, Y . Shu, W. Jiang, and Y . Ha, “A Reliable 8T SRAM for High-Speed Searching and Logic-in-Memory Operations,” IEEE Transactions on Very Large Scale Integration (VLSI) Systems , vol. 30, no. 6, pp. 769–780, 2022

  20. [25]

    A 28-nm Compute SRAM With Bit-Serial Logic/Arithmetic Operations for Programmable In-Memory Vector Computing,

    J. Wang, X. Wang, C. Eckert, A. Subramaniyan, R. Das, D. Blaauw, and D. Sylvester, “A 28-nm Compute SRAM With Bit-Serial Logic/Arithmetic Operations for Programmable In-Memory Vector Computing,” IEEE Journal of Solid-State Circuits , vol. 55, no. 1, pp. 76–86, 2020

  21. [26]

    NUTS-BSNN: A non-uniform time-step binarized spiking neural network with energy-efficient in-memory computing macro,

    V .-N. Dinh, N.-M. Bui, V .-T. Nguyen, D. John, L.-Y . Lin, and Q.-K. Trinh, “NUTS-BSNN: A non-uniform time-step binarized spiking neural network with energy-efficient in-memory computing macro,” Neurocomputing, vol. 560, p. 126838, 2023. [Online]. Available: https://www.scien...

  22. [27]

    CafeHD: A Charge-Domain FeFET-Based Compute-in- Memory Hyperdimensional Encoder with Hypervector Merging,

    T. Li, H. Zhong, J. Wu, T. K ¨ampfe, K. Ni, V . Narayanan, H. Yang, and X. Li, “CafeHD: A Charge-Domain FeFET-Based Compute-in- Memory Hyperdimensional Encoder with Hypervector Merging,” in Design, Automation and Test in Europe Conference and Exhibition (DATE), 2024, pp. 1–6

  23. [28]

    CILP: An Arbitrary-bit Precision All-digital Compute-in-memory Solver for Integer Linear Programming Problems,

    M. Yang, Y . Wang, S. Xie, C.-P. Lo, M. Wang, S. Oruganti, R. Se- hgal, and J. P. Kulkarni, “CILP: An Arbitrary-bit Precision All-digital Compute-in-memory Solver for Integer Linear Programming Problems,” in IEEE Custom Integrated Circuits Conference (CICC) , 2024, pp. 1–2

  24. [29]

    A Scalable and Reconfigurable Bit-Serial Compute-Near-Memory Hardware Accelerator for Solving 2-D/3-D Partial Differential Equations,

    J. Mu, C. Yu, T. T.-H. Kim, and B. Kim, “A Scalable and Reconfigurable Bit-Serial Compute-Near-Memory Hardware Accelerator for Solving 2-D/3-D Partial Differential Equations,” IEEE Journal of Solid-State Circuits, pp. 1–11, 2024

  25. [30]

    Efficient and lightweight in-memory computing architecture for hardware security,

    H. Ajmi, F. Zayer, A. Hadj Fredj, H. Belgacem, B. Mohammad, N. Werghi, and J. Dias, “Efficient and lightweight in-memory computing architecture for hardware security,” Journal of Parallel and Distributed Computing, vol. 190, p. 104898, 2024. [Online]. Available: https://www.sc...

  26. [31]

    Energy Efficient Data Search Design and Optimization Based on a Compact Ferroelectric FET Content Addressable Memory,

    J. Cai, M. Imani, K. Ni, G. L. Zhang, B. Li, U. Schlichtmann, C. Zhuo, and X. Yin, “Energy Efficient Data Search Design and Optimization Based on a Compact Ferroelectric FET Content Addressable Memory,” in Proceedings of the 59th ACM/IEEE Design Automation Conference , ser. DA...

  27. [32]

    BP-SCIM: A Reconfigurable 8T SRAM Macro for Bit-Parallel Searching and Computing In-Memory,

    Y . Chen, J. Mu, H. Kim, L. Lu, and T. T.-H. Kim, “BP-SCIM: A Reconfigurable 8T SRAM Macro for Bit-Parallel Searching and Computing In-Memory,” IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 70, no. 5, pp. 2016–2027, 2023

  28. [33]

    A Reliable and High-Speed 6T Compute-SRAM Design With Dual-Split-VDD Assist and Bitline Leakage Compensation,

    Y . Wang, S. Zhang, Y . Li, J. Chen, W. Zhao, and Y . Ha, “A Reliable and High-Speed 6T Compute-SRAM Design With Dual-Split-VDD Assist and Bitline Leakage Compensation,” IEEE Transactions on Very Large Scale Integration (VLSI) Systems , vol. 31, no. 5, pp. 684–695, 2023

  29. [34]

    A 28 nm Configurable Memory (TCAM/BCAM/SRAM) Using Push-Rule 6T Bit Cell Enabling Logic-in-Memory,

    S. Jeloka, N. B. Akesh, D. Sylvester, and D. Blaauw, “A 28 nm Configurable Memory (TCAM/BCAM/SRAM) Using Push-Rule 6T Bit Cell Enabling Logic-in-Memory,” IEEE Journal of Solid-State Circuits , vol. 51, no. 4, pp. 1009–1021, 2016

  30. [35]

    Two-Direction In-Memory Computing Based on 10T SRAM With Horizontal and Vertical Decoupled Read Ports,

    Z. Lin, Z. Zhu, H. Zhan, C. Peng, X. Wu, Y . Yao, J. Niu, and J. Chen, “Two-Direction In-Memory Computing Based on 10T SRAM With Horizontal and Vertical Decoupled Read Ports,” IEEE Journal of Solid- State Circuits, vol. 56, no. 9, pp. 2832–2844, 2021

  31. [36]

    CAMA: Energy and Memory Efficient Automata Processing in Content-Addressable Memories,

    Y . Huang, Z. Chen, D. Li, and K. Yang, “CAMA: Energy and Memory Efficient Automata Processing in Content-Addressable Memories,” in IEEE International Symposium on High-Performance Computer Archi- tecture (HPCA), 2022, pp. 25–37

  32. [37]

    ABC: An academic industrial-strength verification tool,

    R. Brayton and A. Mishchenko, “ABC: An academic industrial-strength verification tool,” in Computer Aided Verification: 22nd International Conference, CAV , Edinburgh, UK, July 15-19, 2010. Springer, 2010, pp. 24–40

  33. [38]

    Yosys Open SYnthesis Suite,

    C. Wolf, “Yosys Open SYnthesis Suite,” https://yosyshq.net/yosys/

  34. [39]

    The EPFL combi- national benchmark suite,

    L. Amar ´u, P.-E. Gaillardon, and G. De Micheli, “The EPFL combi- national benchmark suite,” in Proceedings of the 24th International Workshop on Logic and Synthesis (IWLS) , no. CONF, 2015

  35. [40]

    A Floating-Point 6T SRAM In-Memory-Compute Macro Using Hybrid- Domain Structure for Advanced AI Edge Chips,

    P.-C. Wu, J.-W. Su, L.-Y . Hong, J.-S. Ren, C.-H. Chien, H.-Y . Chen, C.-E. Ke, H.-M. Hsiao, S.-H. Li, S.-S. Sheu, W.-C. Lo, S.-C. Chang, C.-C. Lo, R.-S. Liu, C.-C. Hsieh, K.-T. Tang, and M.-F. Chang, “A Floating-Point 6T SRAM In-Memory-Compute Macro Using Hybrid- Domain Struc...

  36. [41]

    DDC-PIM: Efficient Algo- rithm/Architecture Co-design for Doubling Data Capacity of SRAM- Based Processing-In-Memory,

    C. Duan, J. Yang, X. He, Y . Qi, Y . Wang, Y . Wang, Z. He, B. Yan, X. Wang, X. Jia, W. Pan, and W. Zhao, “DDC-PIM: Efficient Algo- rithm/Architecture Co-design for Doubling Data Capacity of SRAM- Based Processing-In-Memory,” IEEE Transactions on Computer-Aided Design of Integ...

  37. [42]

    ADRA: Ex- tending Digital Computing-In-Memory With Asymmetric Dual-Row- Activation,

    A. Malhotra, A. K. Saha, C. Wang, and S. K. Gupta, “ADRA: Ex- tending Digital Computing-In-Memory With Asymmetric Dual-Row- Activation,” IEEE Transactions on Circuits and Systems II: Express Briefs, vol. 70, no. 8, pp. 3089–3093, 2023

  38. [43]

    In-Memory Wallace Tree Multipliers Based on Majority Gates Within V oltage-Gated SOT-MRAM Crossbar Arrays,

    Y . Hui, Q. Li, L. Wang, C. Liu, D. Zhang, and X. Miao, “In-Memory Wallace Tree Multipliers Based on Majority Gates Within V oltage-Gated SOT-MRAM Crossbar Arrays,” IEEE Transactions on Very Large Scale Integration (VLSI) Systems , vol. 32, no. 3, pp. 497–504, 2024

  39. [44]

    X-Former: In-Memory Acceleration of Transformers,

    S. Sridharan, J. R. Stevens, K. Roy, and A. Raghunathan, “X-Former: In-Memory Acceleration of Transformers,” IEEE Transactions on Very Large Scale Integration (VLSI) Systems , vol. 31, no. 8, pp. 1223–1233, 2023

  40. [45]

    ADC-Less Reprogrammable RRAM Array Architecture for In-Memory Computing,

    A. Dongre, B. Boro, and G. Trivedi, “ADC-Less Reprogrammable RRAM Array Architecture for In-Memory Computing,” IEEE Trans- actions on Very Large Scale Integration (VLSI) Systems, vol. 31, no. 12, pp. 2053–2060, 2023

  41. [46]

    HARDSEA: Hybrid Analog-ReRAM Clustering and Digital-SRAM In-Memory Computing Accelerator for Dynamic Sparse Self-Attention in Transformer,

    S. Liu, C. Mu, H. Jiang, Y . Wang, J. Zhang, F. Lin, K. Zhou, Q. Liu, and C. Chen, “HARDSEA: Hybrid Analog-ReRAM Clustering and Digital-SRAM In-Memory Computing Accelerator for Dynamic Sparse Self-Attention in Transformer,” IEEE Transactions on Very Large Scale Integration (VL...

  42. [47]

    An RRAM-Based Computing-in-Memory Architecture and Its Application in Accelerating Transformer Inference,

    Z. Lu, X. Wang, M. T. Arafin, H. Yang, Z. Liu, J. Zhang, and G. Qu, “An RRAM-Based Computing-in-Memory Architecture and Its Application in Accelerating Transformer Inference,” IEEE Transactions on Very Large Scale Integration (VLSI) Systems , vol. 32, no. 3, pp. 485–496, 2024

  43. [48]

    Layout-Aware Area Optimization of Transposable STT-MRAM for a Processing-In-Memory System,

    S. Choi, D. Han, C. Choi, and Y . Seo, “Layout-Aware Area Optimization of Transposable STT-MRAM for a Processing-In-Memory System,” IEEE Transactions on Very Large Scale Integration (VLSI) Systems , vol. 32, no. 2, pp. 245–255, 2024

  44. [49]

    CP-SRAM: Charge-Pulsation SRAM Marco for Ultra-High Energy-Efficiency Computing-in-Memory,

    H. Zhang, L. Jiang, J. Wu, T. Chen, J. Liu, W. Kang, and W. Zhao, “CP-SRAM: Charge-Pulsation SRAM Marco for Ultra-High Energy-Efficiency Computing-in-Memory,” in Proceedings of the 59th ACM/IEEE Design Automation Conference , ser. DAC ’22. New York, IEEE TRANSACTIONS ON VERY L...

  45. [50]

    OpenABC-D: A Large-Scale Dataset For Machine Learning Guided Integrated Circuit Synthesis,

    A. B. Chowdhury, B. Tan, R. Karri, and S. Garg, “OpenABC-D: A Large-Scale Dataset For Machine Learning Guided Integrated Circuit Synthesis,” CoRR, vol. abs/2110.11292, 2021. [Online]. Available: https://arxiv.org/abs/2110.11292

  46. [53]

    Design Automation of Series Resonance Clocking in 14-nm FinFETs,

    D. Challagundla, I. Bezzam, and R. Islam, “Design Automation of Series Resonance Clocking in 14-nm FinFETs,” Circuits, Systems, and Signal Processing , Aug. 2023. [Online]. Available: https: //doi.org/10.1007/s00034-023-02458-4

  47. [54]

    Power and Skew Reduction Using Resonance Energy Recycling in FinFET based Wideband Clock Networks,

    D. Challagundla, “Power and Skew Reduction Using Resonance Energy Recycling in FinFET based Wideband Clock Networks,” Master’s thesis, University of Maryland, Baltimore County, 2022

  48. [55]

    Power and skew reduction using resonant energy recycling in 14-nm FinFET clocks,

    D. Challagundla, M. Galib, I. Bezzam, and R. Islam, “Power and skew reduction using resonant energy recycling in 14-nm FinFET clocks,” in 2022 IEEE International Symposium on Circuits and Systems (ISCAS) . IEEE, 2022, pp. 268–272

  49. [56]

    System and methods of reducing wideband series resonant clock skew,

    R. Islam, D. Challagundla, and I. Bezzam, “System and methods of reducing wideband series resonant clock skew,” Oct. 10 2024, US Patent App. 18/627,479

  50. [57]

    Low-Power Resonant Clocking Using Soft Error Robust Energy Recovery Flip-Flops,

    R. Islam, “Low-Power Resonant Clocking Using Soft Error Robust Energy Recovery Flip-Flops,” Journal of Electronic Testing , vol. 34, no. 4, pp. 471–485, jun 2018. [Online]. Available: https://doi.org/10. 1007%2Fs10836-018-5737-6

  51. [59]

    OpenRAM: An open-source memory compiler,

    M. R. Guthaus, J. E. Stine, S. Ataei, B. Chen, B. Wu, and M. Sarwar, “OpenRAM: An open-source memory compiler,” in IEEE/ACM Interna- tional Conference on Computer-Aided Design (ICCAD) , 2016, pp. 1–6

  52. [60]

    Virtual Prototyper (ViPro): An Early Design Space Exploration and Optimization Tool for SRAM Designers,

    S. Nalam, M. Bhargava, K. Mai, and B. H. Calhoun, “Virtual Prototyper (ViPro): An Early Design Space Exploration and Optimization Tool for SRAM Designers,” in Proceedings of the 47th Design Automation Conference, ser. DAC ’10. New York, NY , USA: Association for Computing Mach...

  53. [61]

    Open- SAR: An Open Source Automated End-to-end SAR ADC Compiler,

    M. Liu, X. Tang, K. Zhu, H. Chen, N. Sun, and D. Z. Pan, “Open- SAR: An Open Source Automated End-to-end SAR ADC Compiler,” in IEEE/ACM International Conference On Computer Aided Design (ICCAD), 2021, pp. 1–9

  54. [62]

    AutoDCIM: An Automated Digital CIM Compiler,

    J. Chen, F. Tu, K. Shao, F. Tian, X. Huo, C.-Y . Tsui, and K.-T. Cheng, “AutoDCIM: An Automated Digital CIM Compiler,” in 60th ACM/IEEE Design Automation Conference (DAC) , 2023, pp. 1–6

  55. [63]

    J. M. Rabaey, Digital integrated circuits : a design perspective., 2nd ed., ser. Prentice Hall electronic and VLSI series. Upper Saddle River, N.J: Pearson Education, 2004

  56. [64]

    Bit parallel 6t sram in-memory computing with reconfigurable bit-precision,

    K. Lee, J. Jeong, S. Cheon, W. Choi, and J. Park, “Bit parallel 6t sram in-memory computing with reconfigurable bit-precision,” in 2020 57th ACM/IEEE Design Automation Conference (DAC) , 2020, pp. 1–6

  57. [65]

    A 40-nm cmos multifunctional computing-in-memory (cim) using single-ended disturb-free 7t 1-kb sram,

    C.-C. Wang, L. K. S. Tolentino, C.-Y . Huang, and C.-H. Yeh, “A 40-nm cmos multifunctional computing-in-memory (cim) using single-ended disturb-free 7t 1-kb sram,” IEEE Transactions on Very Large Scale Integration (VLSI) Systems , vol. 29, no. 12, pp. 2172–2185, 2021

  58. [66]

    A fast, reliable and wide-voltage-range in-memory computing architecture,

    W. Simon, J. Galicia, A. Levisse, M. Zapater, and D. Atienza, “A fast, reliable and wide-voltage-range in-memory computing architecture,” in 2019 56th ACM/IEEE Design Automation Conference (DAC) , 2019, pp. 1–6

  59. [67]

    In situ storing 8t sram-cim macro for full-array boolean logic and copy operations,

    Z. Lin, Z. Tong, F. Wang, J. Zhang, Y . Zhao, P. Sun, T. Xu, C. Zhang, X. Li, X. Wu, W. Lu, C. Peng, Q. Zhao, and J. Chen, “In situ storing 8t sram-cim macro for full-array boolean logic and copy operations,” IEEE Journal of Solid-State Circuits , vol. 58, no. 5, pp. 1472–1486, 2023

  60. [68]

    Design of ion-implanted mosfet’s with very small physical dimensions,

    R. Dennard, F. Gaensslen, H.-N. Yu, V . Rideout, E. Bassous, and A. LeBlanc, “Design of ion-implanted mosfet’s with very small physical dimensions,” IEEE Journal of Solid-State Circuits , vol. 9, no. 5, pp. 256–268, 1974. Dhandeep Challagundla (Student Member, IEEE) received h...

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.