Fact-checked by Grok 2 weeks ago

Dennard scaling

Dennard scaling, also known as MOSFET scaling theory, is a foundational principle in semiconductor device physics that outlines how reducing the dimensions of metal-oxide-semiconductor field-effect transistors (MOSFETs) proportionally enhances their speed, power efficiency, and packing density while keeping power dissipation per unit area constant.^[1] Proposed in 1974 by American electrical engineer Robert H. Dennard (1932–2024) and colleagues at IBM in their seminal paper "Design of Ion-Implanted MOSFET's with Very Small Physical Dimensions," the theory assumes that all linear dimensions (such as channel length, width, and gate oxide thickness) are scaled down by a factor \kappa > 1, operating voltages are reduced by $1/\kappa, and substrate doping concentrations are increased by \kappa. This uniform scaling preserves the electric field strengths within the device, preventing excessive short-channel effects and maintaining reliable operation.^[1] Experimental validation in the paper demonstrated these principles using polysilicon-gate MOSFETs scaled to channel lengths as small as 0.5 μm, confirming predictions through fabricated devices.^[1] Under ideal Dennard scaling, circuit performance improves dramatically: propagation delay decreases by a factor of \kappa, enabling higher clock frequencies; power dissipation per circuit drops by \kappa^2, reducing energy use; and active device density rises by \kappa^2, supporting more complex integrated circuits on the same chip area.^[1] These gains aligned closely with Moore's law, which observed the doubling of transistor counts on integrated circuits approximately every two years, by ensuring that power density remained constant as transistors shrank, allowing sustained increases in computing speed without proportional rises in heat generation.^[2] For decades, this synergy drove exponential progress in microprocessor performance, from the 1970s through the early 2000s, powering advancements in personal computing, mobile devices, and data centers.^[3] However, Dennard scaling began to falter around the 90 nm process node in the mid-2000s, primarily due to subthreshold leakage currents that prevented further voltage reductions without severely impacting performance, leading to rising power density and the so-called "power wall."^[4]^[5] Additional challenges included quantum tunneling through thin gate oxides and increased variability in device characteristics at smaller scales, which violated the theory's assumptions of constant mobility and field uniformity.^[6] This breakdown shifted industry focus from single-core frequency scaling to multicore architectures, parallelism, and power management techniques like dynamic voltage scaling, while innovations in materials (e.g., high-k dielectrics) and three-dimensional integration extended scaling benefits beyond classical limits.^[7] Despite these limitations, Dennard scaling remains a cornerstone concept for understanding historical and ongoing semiconductor evolution.^[2]

Core Principles

Statement of Dennard Scaling

Dennard scaling refers to a theoretical framework for uniformly reducing the dimensions of MOSFET transistors while maintaining constant electric fields and power density within the device. As originally proposed, it states that all linear dimensions of the transistor, such as channel length L, width W, and gate oxide thickness t_{ox}, are scaled down by a factor \kappa (where \kappa > 1), the supply voltage V_{DD} and threshold voltage V_t are also reduced by $1/\kappa, and the substrate doping concentration N_a is increased by \kappa. This constant-field scaling ensures that the performance per unit area improves predictably without a rise in power dissipation density.^[1] Under these rules, the drive current per transistor scales by $1/\kappa, the device capacitance scales by $1/\kappa, and the circuit delay time scales by $1/\kappa, leading to faster operation. Power dissipation per device decreases by $1/\kappa^2, but since the device area scales by $1/\kappa^2, the power density remains invariant at a factor of 1. These relationships were derived for ion-implanted MOSFETs designed for digital integrated circuits, demonstrating that miniaturization to channel lengths as small as 0.5 \mum could be achieved while preserving reliability and efficiency.^[1] The original proposal by Robert H. Dennard and colleagues in 1974 enabled the continued scaling of MOSFET technology by outlining how to increase transistor density while keeping power density constant and improving speed through reduced delay times. This approach provided a roadmap for predictable enhancements in circuit performance, such as a power-delay product improvement by a factor of \kappa^3, fostering decades of integrated circuit advancement.

Mathematical Derivation

Dennard scaling relies on the ideal long-channel MOSFET model, assuming uniform reduction of all linear dimensions—such as channel length L, channel width W, and gate oxide thickness t_{ox}—by a factor of $1/\kappa (where \kappa > 1 is the scaling factor), along with proportional scaling of voltages including supply voltage V_{DD} and threshold voltage V_t. This constant-field scaling preserves electric field strengths across the device.^[8] The gate capacitance C_g of a MOSFET is expressed as

C_g = \epsilon_{ox} \frac{W L}{t_{ox}},

where \epsilon_{ox} is the oxide permittivity. Under scaling, W' = W/\kappa, L' = L/\kappa, and t_{ox}' = t_{ox}/\kappa, yielding

C_g' = \epsilon_{ox} \frac{(W/\kappa)(L/\kappa)}{t_{ox}/\kappa} = \epsilon_{ox} \frac{W L}{\kappa^2} \cdot \frac{\kappa}{t_{ox}} = \frac{C_g}{\kappa}.

Thus, the gate capacitance scales inversely with \kappa.^[8] The saturation drain current I_{dsat} in the long-channel square-law regime is given by

I_{dsat} = \frac{\mu \epsilon_{ox}}{2 t_{ox}} \left( \frac{W}{L} \right) (V_{GS} - V_t)^2,

where \mu is carrier mobility (assumed constant). The aspect ratio W/L remains invariant under uniform scaling. Since t_{ox} \propto 1/\kappa, the term \epsilon_{ox}/t_{ox} \propto \kappa, and V_{GS} - V_t \propto 1/\kappa, so (V_{GS} - V_t)^2 \propto 1/\kappa^2. Therefore,

I_{dsat}' \propto \kappa \cdot \frac{1}{\kappa^2} = \frac{1}{\kappa}.

The saturation current scales inversely with \kappa.^[8] Dynamic power dissipation per transistor is P = \alpha C_g V_{DD}^2 f, where \alpha is the activity factor and f is the operating frequency. With C_g \propto 1/\kappa, V_{DD} \propto 1/\kappa (so V_{DD}^2 \propto 1/\kappa^2), and f \propto \kappa (as derived below),

P' \propto \left( \frac{1}{\kappa} \right) \left( \frac{1}{\kappa^2} \right) \kappa = \frac{1}{\kappa^2}.

Equivalently, considering active power as P \propto V_{DD} I_{dsat}, the scaling follows (1/\kappa) \cdot (1/\kappa) = 1/\kappa^2. Power per transistor thus scales as $1/\kappa^2.^[8] The transistor footprint area A \propto W L \propto 1/\kappa^2. Power density is then

PD = \frac{P}{A} \propto \frac{1/\kappa^2}{1/\kappa^2} = 1,

remaining invariant under Dennard scaling. This constancy arises directly from the proportional reduction in power and area.^[8] Circuit speed is characterized by the inverse of propagation delay \tau \propto C_g V_{DD} / I_{dsat}. Substituting the scalings,

\tau' \propto \left( \frac{1}{\kappa} \right) \left( \frac{1}{\kappa} \right) / \left( \frac{1}{\kappa} \right) = \frac{1}{\kappa},

so delay decreases by $1/\kappa, and speed improves by a factor of \kappa.^[8] These relations hold under the ideal assumptions of negligible short-channel effects, constant mobility, and zero subthreshold leakage current, which are valid for feature sizes above approximately 100 nm.^[8]^[9]

Interactions with Scaling Laws

Relation to Moore's Law

Moore's Law, first articulated by Gordon E. Moore in 1965, observes that the number of transistors on an integrated circuit doubles approximately every two years, driven primarily by reductions in feature sizes that enable higher component density at minimum cost.^[10] This empirical trend provided a roadmap for the semiconductor industry, predicting exponential growth in computational capability without specifying underlying physical mechanisms.^[11] Dennard scaling complements Moore's Law by ensuring that as transistor density increases, power density remains constant, allowing voltage and dynamic power to scale down proportionally with linear dimensions.^[12] Under this synergy, shrinking transistors not only doubles their count per chip area but also maintains manageable power levels, preventing thermal bottlenecks that could otherwise halt progress.^[13] This alignment enabled the industry to reap the full benefits of density gains without proportional rises in energy consumption or cooling requirements. The combined scaling typically employs a factor κ ≈ 1.4 per technology generation, corresponding to a roughly 30% linear dimension reduction, which doubles areal density while keeping power dissipation per unit area unchanged in accordance with Dennard's principles.^[14] Historically, from the 1970s through the early 2000s, both laws aligned closely, delivering approximately 2× performance improvement per chip every 18–24 months without a corresponding power increase, as clock frequencies scaled with the shrinking dimensions.^[12] A key outcome of this joint scaling is that computations per joule improve by a factor of κ³ per generation, enhancing overall energy efficiency at the system level.^[15]^[9] While Moore's Law emerged as an observational pattern of industry trends, Dennard scaling represents a physics-based prediction grounded in MOSFET device behavior, providing the theoretical foundation that made sustained density scaling feasible.^[16]

Impact on Computing Performance

Dennard scaling enabled significant improvements in transistor performance by reducing the delay time per transistor by a factor of $1/\kappa, where \kappa is the scaling factor greater than 1 for each technology generation. This reduction in delay directly allowed clock frequencies to increase proportionally with \kappa, facilitating a transition from megahertz-range processors in the 1970s to gigahertz speeds by the early 2000s. For instance, Intel's processors evolved from the 8080 at 2 MHz in 1974 to the Pentium 4 exceeding 3 GHz by 2004, demonstrating how scaling sustained rapid single-core performance gains over decades.^[9]^[17] In terms of energy efficiency, the energy per transistor operation scaled down by $1/\kappa^3 under ideal Dennard conditions, as capacitance scaled with $1/\kappa and voltage with $1/\kappa. This contributed to Koomey's Law, which observed that the number of computations per joule of energy roughly doubled every 1.57 years from the 1940s through the 2000s, reflecting orders-of-magnitude improvements in electrical efficiency driven by scaling. From 1974 to 2004 specifically, performance per watt advanced by several orders of magnitude, correlating closely with adherence to Dennard principles and enabling more complex workloads within constrained power budgets.^[18]^[14] At the chip level, constant power density meant that total chip power P_{\text{chip}} scaled linearly with transistor count, which followed Moore's Law as P_{\text{chip}} \propto 2^{t/\tau} with \tau \approx 2 years, allowing power growth to remain manageable through conventional cooling until the mid-2000s. This scaling supported the proliferation of portable computing in the 1990s and 2000s, as efficient single-core processors powered laptops and mobile devices without excessive heat or battery drain. While Dennard scaling primarily drove transistor-level gains, it indirectly facilitated parallel advances in memory density and interconnect speeds, amplifying overall system performance.^[14]^[9]

Historical Evolution and Limitations

Development and Historical Context

Dennard scaling was proposed by the late Robert H. Dennard (1932–2024), an IBM researcher, in the 1974 paper "Design of Ion-Implanted MOSFET's with Very Small Physical Dimensions," co-authored with colleagues at IBM's Thomas J. Watson Research Center.^[19] This work emerged during early experimentation with complementary metal-oxide-semiconductor (CMOS) technology, building on the invention of the metal-oxide-semiconductor field-effect transistor (MOSFET) in the 1960s by Mohamed Atalla and Dawon Kahng at Bell Laboratories.^[20] It addressed the scaling limitations of bipolar transistors, which suffered from high power dissipation and heat generation that hindered dense integration in logic and memory circuits.^[21] Developed amid IBM's efforts to advance high-density digital integrated circuits, such as dynamic random-access memory (DRAM), the theory was motivated by the need to achieve higher transistor integration without increasing power or heat density, enabling reliable performance in miniaturized devices.^[22] The scaling approach relied on the planar MOSFET structure, where all device dimensions, including gate length, oxide thickness, and junction depths, were reduced proportionally while maintaining constant electric fields through adjusted voltages and doping concentrations.^[1] Dennard scaling gained widespread adoption in the 1980s and 1990s as CMOS technology became the dominant fabrication process for very-large-scale integration (VLSI) chips, replacing earlier NMOS and bipolar approaches due to its low power consumption and scalability.^[23] It formed a foundational element of VLSI design rules, guiding systematic dimension reductions in semiconductor manufacturing.^[24] Early validations occurred through IBM experiments fabricating polysilicon-gate MOSFETs with channel lengths ranging from 0.5 to 10 μm, using techniques like electron-beam lithography, which confirmed the theory's predictions for threshold voltage and short-channel effects.^[1] These results, targeting features around 1 μm, demonstrated improved circuit speed and reduced power per device, influencing subsequent industry roadmaps such as the International Technology Roadmap for Semiconductors (ITRS).^[25] The principles were quickly embraced as a systematic guide for MOSFET evolution, shaping decades of semiconductor progress.^[9]

Breakdown Around 2006

Dennard scaling held effectively through the 90 nm technology node, which entered production around 2004-2005, but began to falter as feature sizes dropped below the 65 nm node in 2006-2007.^[26]^[9] At these scales, the ideal assumptions of uniform voltage reduction and constant power density could no longer be maintained, marking the transition away from classical MOSFET scaling.^[9] A key factor in this breakdown was the surge in subthreshold leakage current, governed by the exponential relationship

I_{\text{leak}} \propto \exp\left( -\frac{V_t}{n V_{\text{th}}} \right),

where V_t is the threshold voltage, n is the subthreshold swing coefficient (ideally 1 but typically 1.3-2 in practice), and V_{\text{th}} = kT/q \approx 26 mV at room temperature is the thermal voltage.^[9] Threshold voltage scaling stalled below 0.3-0.4 V, as further reductions triggered unacceptably high off-state leakage, rising from less than $10^{-10} A/μm in earlier nodes to over $10^{-7} A/μm by the mid-2000s.^[9] Short-channel effects, including drain-induced barrier lowering (DIBL) and carrier velocity saturation, further exacerbated the issue by disrupting the uniform electric field assumption central to Dennard scaling, requiring higher channel doping that degraded mobility and increased junction leakage.^[9] The inability to scale supply voltage V_{DD}, which plateaued near 1 V (e.g., 1.1 V at the 65 nm node), compounded these problems.^[26] Under Dennard rules, dynamic power follows P \propto C V_{DD}^2 f, where C is capacitance and f is frequency; with V_{DD} fixed, power grew nearly linearly with frequency increases and the slower-than-expected capacitance reduction due to fringing fields.^[9] This led to escalating power density, forming the "power wall," as total chip power surpassed 100-150 W—limits for reliable air cooling in desktop and server systems.^[26] By 2007, high-performance MPU power density reached 0.64 W/mm², a level that persisted amid stalled frequency gains.^[26] Empirical evidence underscored this shift: microprocessor clock frequencies peaked at 3-4 GHz, exemplified by the Intel Pentium 4's 3.8 GHz model in 2004-2005, which consumed up to 115 W and highlighted thermal bottlenecks.^[27] The International Technology Roadmap for Semiconductors (ITRS) 2006 update explicitly flagged the end of classical scaling, citing leakage and power constraints as barriers to sustained performance gains.^[26] Additional contributors included heightened variability in nanoscale doping, which caused inconsistent threshold voltages across transistors, and quantum mechanical tunneling through gate oxides thinner than 2 nm (e.g., 1.2 nm equivalent oxide thickness at 65 nm), introducing significant gate leakage currents.^[9] These effects collectively terminated the era of predictable, density-constant power scaling.^[9]

Post-Dennard Implications

The breakdown of Dennard scaling around 2006 prompted an immediate shift in processor design toward multicore architectures to harness continued transistor density improvements through parallelism rather than relying on single-core clock speed increases. For instance, Intel introduced the Core 2 Duo processor in 2006, which emphasized dual-core configurations to deliver performance gains while managing power constraints more effectively than prior single-core designs.^[28]^[29] This transition allowed the industry to sustain computational progress by distributing workloads across multiple cores, mitigating the inability to scale voltage and frequency proportionally with feature size. A key consequence has been the emergence of "dark silicon," where power and thermal limits prevent the simultaneous activation of all transistors on a chip, resulting in utilization rates often below 50% in modern system-on-chips (SoCs). This phenomenon arises because shrinking transistors increases leakage currents and dynamic power density, forcing designers to power down portions of the die to avoid exceeding thermal budgets.^[30]^[31] To address these challenges, new transistor paradigms like FinFETs, introduced by Intel in 2011 at the 22 nm node, and gate-all-around FETs (GAAFETs) in the 2020s have partially restored scaling benefits by improving electrostatic control and reducing short-channel effects, though they fall short of full Dennard proportionality due to persistent power density issues.^[32]^[33] Complementing these, 3D stacking via chiplets enables higher integration density without further planar dimension shrinks, as seen in advanced packaging standards that support vertical interconnects for enhanced bandwidth and efficiency.^[34]^[35] Post-breakdown efficiency trends reflect a slowdown in Koomey's Law, which historically described computations per joule doubling roughly every 1.5 years; since around 2006, improvements have decelerated to about every 2.5 years, driven by the decoupling of power and performance scaling.^[36] Despite this, gains persist through specialized accelerators like GPUs and tensor processing units (TPUs), which optimize for parallel workloads such as AI training, achieving higher energy efficiency for targeted tasks compared to general-purpose CPUs.^[37] Recent developments from 2020 to 2025 have intensified beyond-CMOS research, including 2D materials like transition metal dichalcogenides for low-power devices and neuromorphic computing architectures that mimic brain-like efficiency to bypass von Neumann bottlenecks. For instance, in January 2025, imec proposed integrating 2D materials like transition metal dichalcogenides into CFET architectures to extend logic scaling, while Intel's 18A process, featuring RibbonFET GAAFETs, began high-volume production in the second half of 2025, aiming to improve performance per watt by up to 15%.^[38]^[39]^[40]^[41] The International Roadmap for Devices and Systems (IRDS) highlights a "More than Moore" era focused on heterogeneous integration, combining diverse technologies for system-level performance rather than pure transistor scaling.^[42] A 2021 analysis by Koomey underscores continued, albeit decelerated, efficiency improvements in data processing and storage, emphasizing the role of architectural innovations in sustaining progress.^[43] Looking ahead, transistor scaling faces fundamental limits below 1 nm, where quantum effects like tunneling degrade reliability, prompting a pivot toward software optimizations—such as advanced algorithms and compiler techniques—to extract more performance from existing hardware, alongside energy harvesting methods like photovoltaic integration in IoT devices to reduce external power demands.^[44] These strategies aim to extend computational capabilities in an era where hardware scaling alone can no longer drive exponential gains.^[45]

References

[1]
[PDF] Design of ion-implanted MOSFET's with very small physical ...
Applying this scaling approach to a properly designed conventional-size. MOSFET shows that a 200-A gate insulator is required if the channel length is to be ...
[2]
Dennard's Law - Semiconductor Engineering
Dennard's Law states that as the dimensions of a device go down, so does power consumption. While this held, smaller transistors ran faster, used less power, ...<|control11|><|separator|>
[3]
1974: Scaling of IC Process Design Rules Quantified
IBM researcher Robert Dennard's paper on process scaling on MOS memories accelerates a global race to shrink physical dimensions and manufacture ever more ...
[4]
[PDF] Is Dark Silicon Useful?
Jun 3, 2012 · Due to the breakdown of Dennardian Scaling, multicore chips will not be able to scale with die area; the fraction of a chip that can be ...
[5]
[PDF] 1. CMOS Scaling - nanoHUB
Classic Scaling. Nikonov 1. CMOS. 10. R. Dennard, IEEE JSSC, 1974. Device or Circuit Parameter Scaling Factor. Device dimension tox, L, W. 1/κ. Doping ...
[6]
Classic Moore's Law Scaling Challenges Demand New Ways to ...
May 23, 2022 · Dennard scaling ended when it became impossible to push single-core chips to higher speeds without creating excessive heat. As a result ...
[7]
https://www.appliedmaterials.com/us/en/blog/blog-posts/classic-moores-law-scaling-challenges-demand-new-ways-wire-and-integrate-chips.html
[8]
[PDF] A 30 Year Retrospective on Dennard's MOSFET Scaling Paper
Dennard, et al., “Design of ion-implanted. MOSFETs with very small physical dimensions,”. IEEE Journal of Solid State Circuits, vol. SC-9, no. 5, pp. 256-268, ...Missing: original | Show results with:original
[9]
[PDF] Cramming more components onto integrated circuits
Electronics, Volume 38, Number 8, April 19, 1965. The future of integrated electronics is the future of electron- ics itself. The advantages of integration ...
[10]
Exponential Laws of Computing Growth - Communications of the ACM
Jan 1, 2017 · Moore's Law is one small component in an exponentially growing planetary computing ecosystem. By Peter J. Denning and Ted G. Lewis
[11]
Moore's Law Milestones - IEEE Spectrum
Apr 30, 2015 · Robert H. Dennard and colleagues at IBM describe how transistors can scale reliably to smaller dimensions, the essence of Moore's Law. The ...
[12]
https://spectrum.ieee.org/moores-law-milestones
[13]
Chapter: Appendix E: Dennard Scaling and Implications
“In a classic 1974 paper, reprinted in Appendix D, Robert Dennard et al. showed that the MOS transistor has a set of very convenient scaling properties.10 ...
[14]
[PDF] Moore's Law in CLEAR Light - arXiv
size, power and cost for scaling of transistors (Dennard Scaling and Moore's second law) became a dominant factor during the next period (1965-1977) ...
[15]
Intel® Microprocessor Quick Reference Guide - Year
2005. November, 2005. Intel® Pentium® 4 Processor 662 and 672 3.6 - 3.8 GHz. Dual Core Intel® Xeon® Processor 7000 2.666 - 3 GHz. October, 2005. Dual Core Intel ...Missing: clock speeds
[16]
Implications of Historical Trends in the Electrical Efficiency of ...
Mar 29, 2010 · The electrical efficiency of computation has doubled roughly every year and a half for more than six decades, a pace of change comparable to that for computer ...
[17]
Design of Ion-Implanted MOSFET's With Very Small Physical ...
Jan 1, 1974 · Design of Ion-Implanted MOSFET's With Very Small Physical Dimensions for IEEE JSSC by Robert H. Dennard et al.
[18]
1960: Metal Oxide Semiconductor (MOS) Transistor Demonstrated
In 1960 Karl Zaininger and Charles Meuller fabricated an MOS transistor at RCA and C.T. Sah of Fairchild built an MOS-controlled tetrode. Fred Heiman and Steven ...Missing: CMOS | Show results with:CMOS
[19]
The Impact of Dennard's Scaling Theory - IEEE - YUMPU
Feb 9, 2013 · MOSFET would someday replace the bipolar transistor in<br />. high-performance logic and memory applications.<br />. Driving the Demonstration ...
[20]
The Ripple Effects of Robert H. Dennard, Inventor of DRAM and ...
Feb 9, 2022 · MOSFET scaling relationships from Dennard's seminal 1974 paper entitled “Design of Ion-Implanted MOSFET's with Very Small Physical Dimensions.Missing: original | Show results with:original
[21]
The History of the Integrated Circuit - AnySilicon
1980s: CMOS technology becomes the dominant technology for integrated circuits. Present: Continuous advancements in chip density, driven by Moore's Law ...
[22]
[PDF] Mosfet Scaling-the Driver of VLSI Technology - CS.HUJI
This is an introduction to the Classic Paper on MOSFET scaling by R. Dennard et al., “Design of Ion-Implanted MOSFET's with. Very Small Physical Dimensions ...
[23]
IRDS 2022 Executive Summary
(NTRS) and the International Technology Roadmap for Semiconductors (ITRS)) concentrated on forecasting the rate of transistor ... Dennard's scaling guidelines led ...
[24]
[PDF] Executive Summary - Semiconductor Industry Association
MPU Physical Gate Length targets are unchanged from the 2003 ITRS through the 2006 ITRS Update, but also included are the complete set of annualized Long ...
[25]
[PDF] Intel(R) Pentium(R) 4 Processor on 90 nm Process Datasheet
The voltage set by the VID pins is the maximum voltage allowed by the processor. A minimum voltage is provided in Table 9 and changes with frequency. This ...
[26]
Intel Unveils World's Best Processor
Jul 27, 2006 · The Core 2 Duo desktop processor is an energy-efficient marvel, packing 291 million transistors yet consuming 40 percent lower power, while ...<|separator|>
[27]
Power Challenges May End the Multicore Era
Feb 1, 2013 · The recent shift to multicore designs, which was partly a response to the end of Dennard scaling, aimed to continue proportional performance ...
[28]
Dark Silicon - Semiconductor Engineering
Dark silicon is a method of conserving power by powering down non-operational transistors, which are a fraction of a chip's area.
[29]
(PDF) A Perspective on Dark Silicon - ResearchGate
In order to avoid too high power dissipation, a certain part of the chip needs to remain inactive; the inactive part is termed dark silicon [5]. Hence, we have ...
[30]
FinFET Scaling Reaches Thermal Limit - Semiconductor Engineering
May 12, 2016 · Dennard was working as an IBM researcher. He introduced the idea that MOSFETs would continue to work as voltage-controlled switches in ...Missing: 2011 GAAFET 2020s partial
[31]
Comprehensive Review of FinFET Technology: History, Structure ...
Sep 25, 2024 · This comprehensive review delves into the intricacies of FinFET fabrication, exploring historical development, classifications, and cutting-edge ideas.Missing: 2011 2020s Dennard
[32]
Partitioning In The Chiplet Era - Semiconductor Engineering
Oct 3, 2024 · Shrinking features to add more compute density into a planar SoC is no longer cost-effective for many applications, and it limits the ...
[33]
High-performance, power-efficient three-dimensional system-in ...
Feb 19, 2024 · Universal chiplet interconnect express (UCIe) is an open industry standard interconnect offering high-bandwidth, low-latency, power-efficient ...
[34]
Electrical efficiency of computing - AI Impacts
In 2015 Koomey and Naffziger reported in IEEE Spectrum that Koomey's law began to slow down in around 2000 and by 2015, electrical efficiency was taking 2.5 ...
[35]
There's plenty of room at the Top: What will drive computer ... - Science
Jun 5, 2020 · Google has developed a tensor-processing unit (TPU) (52) specifically designed for deep learning, embracing special-purpose processing and ...
[36]
https://aiimpacts.org/electrical-efficiency-of-computing/
[37]
A neuromorphic processor with on-chip learning for beyond-CMOS ...
Jul 11, 2025 · We present a mixed-signal neuromorphic architecture designed to explore the integration of on-chip learning circuits and novel two- and three-terminal devices.
[38]
IRDS: Device Characteristics for More Moore and Considerations for ...
Aug 5, 2024 · The International Roadmap of Devices and Systems (IRDS) presents device characteristics for extending. Moore's Law with “More Moore.” Today's ...
[39]
Avoiding pitfalls assessing the Internet's energy and carbon impacts
Jul 21, 2021 · Sometimes efficiency increases will exceed growth in service demand, for example, in the case of Cogent, a major network provider that has ...
[40]
[PDF] Energy Efficiency Scaling for Two Decades Research and ...
... (Dennard scaling), leading to more than doubling energy efficiency biennially. By 2005, however, this biennial efficiency doubling began to slow markedly as ...
[41]
How we get to 1 nanometer chips and beyond - IBM Research
Dec 5, 2022 · We have identified two major breakthroughs that we believe will lead us down a path to designing computer chip nodes targeting for 1nm and beyond.Missing: outlook software optimization harvesting