Thermal Engineering at the Limits of the CMOS Era

Valavala, Krishna; Coulson, Keith D.; Rajagopal, Manjunath C.; Gelda, Dhruv; Sinha, Sanjiv

doi:10.1016/b978-0-12-812311-9.00004-9

Cited by 3 publications

(2 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It has successfully supported fiber capacity growth over the past decade [3]- [6]. However, in handling the data rates in state-of-the-art optical communication systems, the power dissipation of 7 nm complementary metal-oxidesemiconductor (CMOS) DSP chips for 800-Gigabit Ethernet has already approached the maximum thermal dissipation capacity of modern packaging technologies [7], [8]. The problem of surpassing the thermal capacity has necessitated a careful compromise in the complexity of DSP algorithms to ensure that power consumption is within an acceptable limit.…”

Section: Introductionmentioning

confidence: 99%

Multi-Wavelength Photonic Neuromorphic Computing for Intra and Inter-Channel Distortion Compensations in WDM Optical Communication Systems

Wang

Lima

Shastri

et al. 2023

IEEE J. Select. Topics Quantum Electron.

View full text Add to dashboard Cite

DSP (digital signal processing) has been widely applied in optical communication systems to mitigate various signal distortions and has become one of the key technologies that have sustained data traffic growth over the past decade. However, the strict energy budget of application-specific integrated circuit-based DSP chips has prevented the deployment of some powerful but computationally costly DSP algorithms in real applications. As a result, fiber nonlinearity-induced signal distortions impede fiber communications systems, especially in wavelength-division multiplexed (WDM) transmission systems.To solve these challenges in DSP, there has been a surge of interest in implementing neural networks-based signal processing using photonics hardware (i.e., photonic neural networks). Photonic neural networks promise to break performance limitations in electronics and gain advantages in bandwidth, latency, and power consumption in solving intellectual tasks that are unreachable by conventional digital electronic platforms. This work proposes a photonic recurrent neural network (RNN) capable of simultaneously resolving dispersion and both intra-and inter-channel fiber nonlinearities in multiple WDM channels in the photonic domain, for the first time to our best knowledge. Furthermore, our photonic RNN can directly process optical WDM signals in the photonic domain, avoiding prohibitive energy consumption and speed overhead in analog to digital converters (ADC). Our proposed photonic RNN is fully compatible with mature silicon photonic fabrications. We demonstrate in simulation that our photonic RNN can process multiple WDM channels simultaneously and achieve a reduced bit error rate compared to typical DSP algorithms for all WDM channels in a pulse-amplitude modulation 4-level (PAM4) transmission system, thanks to its unique capability to address inter-channel fiber nonlinearities. In addition to signal quality performance, the proposed system also promises to significantly reduce the power consumption and the latency compared to the state-of-the-art DSP chips, according to our power and latency analysis.

show abstract

Section: Introductionmentioning

confidence: 99%

Multi-Wavelength Photonic Neuromorphic Computing for Intra and Inter-Channel Distortion Compensations in WDM Optical Communication Systems

Wang

Lima

Shastri

et al. 2023

IEEE J. Select. Topics Quantum Electron.

View full text Add to dashboard Cite

show abstract

“…For instance, changing the pressure applied across two contacting blocks by 0-3 MPa can change the measured TIR by an order of ~10 3 [36], [37]. On the other hand, interfaces used in electronics packaging [10], [38], [39], aviation [8] [9], and other commercial applications are often bonded using adhesives or welds. Adhesively bonded interfaces are typically cured at high temperatures and pressures (~190°C and 2 MPa) [40], which also improves the physical adsorption and diffusion between the adhesive and adherent [41].…”

Section: Introductionmentioning

confidence: 99%

Intrinsic thermal interfacial resistance measurement in bonded metal–polymer foils

Rajagopal

Man

Agrawal

et al. 2020

Review of Scientific Instruments

Self Cite

View full text Add to dashboard Cite

Heat conduction through bonded metal-polymer interfaces often limits the overall heat transfer in electronics packaging, batteries, and heat recovery systems. To design the thermal circuit in such systems, it is essential to measure the thermal interfacial resistance (TIR) across ~1-100 μm junctions. Previously reported TIR of metal-polymer junctions utilize ASTM E1530-based twoblock systems that measure the TIR by applying pressure across the interface through external heating and cooling blocks. Here, we report a novel modification of the ASTM-E1530 technique that employs integrated heaters and sensors to provide an intrinsic TIR measurement of an adhesively bonded metal-polymer junction. We design the measurement technique using finite element simulations to either passively suppress or actively compensate the lateral heat diffusion through the polymer, which can minimize the systematic error to ≲5%. Through proof-of-concept experiments, we report the TIR of metal-polymer interfaces made from DuPont's Pyralux doubleside copper-clad laminates, commonly used in flexible printed circuit boards. Our TIR measurement errors are <10%. We highlight additional sources of errors due to non-idealities in the experiment and discuss possible ways to overcome them. Our measurement technique is also applicable to interfaces that are electrically insulating such as adhesively-joined metal-metal junctions and sputter-coated or welded metal-polymer junctions. Overall, the technique is capable of measuring TIR ≳10 -5 m 2 KW -1 in bonded metal-polymer foils, and can be tailored for in situ measurements in flexible electronics, circuit packaging, and other hybrid metal-polymer systems.

show abstract

Quantum computers, quantum computing, and quantum thermodynamics

Cleri

2024

Front. Quantum Sci. Technol.

View full text Add to dashboard Cite

Quantum thermodynamics aims to extend standard thermodynamics and non-equilibrium statistical physics to systems with sizes well below the thermodynamic limit. It is a rapidly evolving research field that promises to change our understanding of the foundations of physics, while enabling the discovery of novel thermodynamic techniques and applications at the nanoscale. Thermal management has turned into a major obstacle in pushing the limits of conventional digital computers and could also represent a crucial issue for quantum computers. The practical realization of quantum computers with superconducting loops requires working at cryogenic temperatures to eliminate thermal noise, and ion-trap qubits also need low temperatures to minimize collisional noise. In both cases, the sub-nanometric sizes also bring about the thermal broadening of the quantum states; and even room-temperature photonic computers eventually require cryogenic detectors. A number of thermal and thermodynamic questions, therefore, take center stage, such as quantum re-definitions of work and heat, thermalization and randomization of quantum states, the overlap of quantum and thermal fluctuations, and many others, even including a proper definition of temperature for the small open systems constantly out of equilibrium that are the qubits. This overview provides an introductory perspective on a selection of current trends in quantum thermodynamics and their impact on quantum computers and quantum computing, with language that is accessible to postgraduate students and researchers from different fields.

show abstract

Thermal Engineering at the Limits of the CMOS Era

Cited by 3 publications

References 60 publications

Multi-Wavelength Photonic Neuromorphic Computing for Intra and Inter-Channel Distortion Compensations in WDM Optical Communication Systems

Multi-Wavelength Photonic Neuromorphic Computing for Intra and Inter-Channel Distortion Compensations in WDM Optical Communication Systems

Intrinsic thermal interfacial resistance measurement in bonded metal–polymer foils

Quantum computers, quantum computing, and quantum thermodynamics

Contact Info

Product

Resources

About