scispace - formally typeset
Search or ask a question
Journal ArticleDOI

Single-chip microprocessor that communicates directly using light

TL;DR: This demonstration could represent the beginning of an era of chip-scale electronic–photonic systems with the potential to transform computing system architectures, enabling more powerful computers, from network infrastructure to data centres and supercomputers.
Abstract: An electronic–photonic microprocessor chip manufactured using a conventional microelectronics foundry process is demonstrated; the chip contains 70 million transistors and 850 photonic components and directly uses light to communicate to other chips. The rapid transfer of data between chips in computer systems and data centres has become one of the bottlenecks in modern information processing. One way of increasing speeds is to use optical connections rather than electrical wires and the past decade has seen significant efforts to develop silicon-based nanophotonic approaches to integrate such links within silicon chips, but incompatibility between the manufacturing processes used in electronics and photonics has proved a hindrance. Now Chen Sun et al. describe a 'system on a chip' microprocessor that successfully integrates electronics and photonics yet is produced using standard microelectronic chip fabrication techniques. The resulting microprocessor combines 70 million transistors and 850 photonic components and can communicate optically with the outside world. This result promises a way forward for new fast, low-power computing systems architectures. Data transport across short electrical wires is limited by both bandwidth and power density, which creates a performance bottleneck for semiconductor microchips in modern computer systems—from mobile phones to large-scale data centres. These limitations can be overcome1,2,3 by using optical communications based on chip-scale electronic–photonic systems4,5,6,7 enabled by silicon-based nanophotonic devices8. However, combining electronics and photonics on the same chip has proved challenging, owing to microchip manufacturing conflicts between electronics and photonics. Consequently, current electronic–photonic chips9,10,11 are limited to niche manufacturing processes and include only a few optical devices alongside simple circuits. Here we report an electronic–photonic system on a single chip integrating over 70 million transistors and 850 photonic components that work together to provide logic, memory, and interconnect functions. This system is a realization of a microprocessor that uses on-chip photonic devices to directly communicate with other chips using light. To integrate electronics and photonics at the scale of a microprocessor chip, we adopt a ‘zero-change’ approach to the integration of photonics. Instead of developing a custom process to enable the fabrication of photonics12, which would complicate or eliminate the possibility of integration with state-of-the-art transistors at large scale and at high yield, we design optical devices using a standard microelectronics foundry process that is used for modern microprocessors13,14,15,16. This demonstration could represent the beginning of an era of chip-scale electronic–photonic systems with the potential to transform computing system architectures, enabling more powerful computers, from network infrastructure to data centres and supercomputers.

Summary (1 min read)

Jump to:  and [Methods]

Methods

  • The key chip characteristics are summarized in Extended Data Table 1 .
  • Afterwards, the chips are placed in a chamber which supplies XeF 2 gas to isotropically etch the silicon substrate, removing it as the volatile product SiF 4 .
  • The authors use lensed fibers available from Oz Optics with a spot size of 5 µm and a working distance of 26 µm to couple light into the vertical grating couplers through the chip backside (after substrate removal).
  • Programs are compiled from C source code using a gcc-based C compiler targeted for the RISC-V ISA.

Did you find this useful? Give us your feedback

Content maybe subject to copyright    Report

UC Berkeley
UC Berkeley Previously Published Works
Title
Single-chip microprocessor that communicates directly using light.
Permalink
https://escholarship.org/uc/item/4dh1v4px
Journal
Nature, 528(7583)
ISSN
0028-0836
Authors
Sun, Chen
Wade, Mark T
Lee, Yunsup
et al.
Publication Date
2015-12-01
DOI
10.1038/nature16454
Copyright Information
This work is made available under the terms of a Creative Commons Attribution-
NonCommercial-NoDerivatives License, availalbe at
https://creativecommons.org/licenses/by-nc-nd/4.0/
Peer reviewed
eScholarship.org Powered by the California Digital Library
University of California

1
Single-Chip Microprocessor with Integrated Photonic I/O
Chen Sun
*1,2
, Mark T. Wade
*3
, Yunsup Lee
*1
, Jason S. Orcutt
*2,4
, Luca Alloatti
2
, Michael S.
Georgas
2
, Andrew S. Waterman
1
, Jeffrey M. Shainline
3,5
, Rimas R. Avizienis
1
, Sen Lin
1
, Benjamin
R. Moss
2
, Rajesh Kumar
3
, Fabio Pavanello
3
, Amir H. Atabaki
2
, Henry M. Cook
1
, Albert J. Ou
1
,
Jonathan C. Leu
2
, Yu-Hsin Chen
2
, Krste Asanović
1
, Rajeev J. Ram
2
, Miloš A. Popović
3
, Vladimir
M. Stojanović
1
*
Contributed equally to this work
1
University of California, Berkeley, Berkeley, CA
2
Massachusetts Institute of Technology, Cambridge, MA
3
University of Colorado, Boulder, Boulder, CO
4
Now at IBM T.J. Watson Research Center, Yorktown Heights, NY
5
Now at National Institute of Science and Technology, Boulder, CO
Data transport across short-reach electrical wires is both bandwidth-density and power-
density-limited, creating a performance bottleneck for semiconductor microchips in modern
computer systems, from mobile phones to large-scale datacenters. Optical communications
based on chip-scale electronic-photonic systems
1–4
enabled by silicon-based nanophotonic
devices
5
can overcome these limitations
6–8
. However, combining electronics and photonics on
the same chip has proved challenging due to microchip manufacturing conflicts between
electronics and photonics. Consequently, current electronic-photonic chips
9–11
are limited to
niche manufacturing processes and integrate only a few optical devices alongside simple

2
circuits. Here we report an electronic-photonic system-on-chip (SoC) integrating over 70
million transistors and 850 photonic components that work in concert to provide logic,
memory, and interconnect functions, realizing a microprocessor chip that can optically
communicate directly to the outside world for the first time. To integrate electronics and
photonics at this scale, we adopt a zero-change approach to the integration of photonics.
Instead of developing a custom process to enable the fabrication of photonics
12
, which
complicates or eliminates the possibility of integration with state-of-the-art transistors at
large scale and at high yield, we design optical devices directly within a standard
microelectronics foundry process used for modern microprocessors
13
(Cell
14
, BlueGene/Q
15
,
Power7
16
, etc.). We expect that this demonstration signals the beginning of an era of
electronic-photonic SoCs with potential for transformative impact on computing system
architecture, enabling a leap to new kinds of more powerful computers, from network
infrastructure to datacenters and supercomputers.
The electro-optic SoC (Figure 1) contains a dual-core RISC-V instruction set architecture
17
(ISA)
microprocessor and an independent 1 MB bank of static random access memory used for memory.
The on-chip electro-optic transceivers for data input/output (I/O) enable both the microprocessor
and the memory to communicate directly to off-chip components using light, without the need for
separate chips or components to host the optical devices. The chip was fabricated in a commercial
high-performance 45 nm complementary metal-oxide semiconductor (CMOS) silicon-on-insulator
(SOI) process
18
. No changes to the foundry process were necessary to accommodate photonics and
all optical devices were designed to comply with the native process manufacturing rules. This
zero-change integration enables high-performance transistors on the same chip as optics, reuse of
all existing designs in the process, compatibility with electronics design tools, and manufacturing

3
in an existing high-volume foundry.
The process includes a crystalline Si (c-Si) layer which is patterned to form both the body of
electronic transistors and the core of optical waveguides. A thin buried oxide (BOX) layer
separates the c-Si layer from the silicon handle wafer (Extended Data Figure 1). As the BOX is
<200 nm thick, light propagating in c-Si waveguides will evanescently leak into the silicon handle
wafer, resulting in high waveguide loss. To resolve this, we perform selective substrate removal on
the chips after electrical packaging to etch away the silicon handle under regions with optical
devices (Extended Data Figure 2). We leave the silicon handle intact under the microprocessor and
memory (which dissipate the most power) to allow a heat sink to be contacted, if necessary.
Substrate removal has a negligible impact on the electronics
13
and the processor is completely
functional even with a fully-removed substrate.
Silicon-germanium (SiGe) is present, though in low germanium mole fractions, in advanced
CMOS processes for enhancing hole mobility and transistor performance via compressive strain
engineering of p-channel transistors
18
. Selecting an 1180 nm wavelength band for the optical
channel enables use of photodetectors (PDs) built using this SiGe
19
. Silicon is transparent at 1180
nm and no adverse effects are observed. At these wavelengths, the optical propagation loss in
silicon strip waveguides is 4.3 dB/cm (losses at industry-standard wavelengths of 1300 nm and
1550 nm are 3.7 dB/cm and 4.6 dB/cm, respectively
13
). The receiver circuit
20
resolves photocurrent
produced by the illuminated PD into digital ones and zeros. The receiver sensitivity in optical
modulation amplitude (OMA) is 5 dBm for a better than 10
12
bit-error-rate.
The electro-optic transmitter consists of an electro-optic modulator and its electronic driver. The

4
modulator is a 10 µm diameter silicon microring resonator, coupled to a waveguide. We dope the
structure with the n-well and p-well implants used for transistors to form radially extending p-n
junctions, interleaved along the azimuthal dimension
21, 22
, taking the form of a “spoked ring”. The
ring exhibits a sharp, notched filter optical transmission response, with a stop-band at the ring’s
resonant wavelength (λ
0
). Applying a negative voltage across the junctions depletes the ring of free
carriers (electron and hole concentrations), while a small positive voltage refills the carriers. A
change in carrier concentration influences the ring waveguide’s index of refraction through the
carrier plasma dispersion effect
23
which, in turn, shifts λ
0
. Electro-optic modulation (on-off keying)
is achieved by changing the voltage applied across the junction to move the λ
0
stop-band in and out
of the laser wavelength (λ
L
). The modulator has a loaded quality factor of approximately 10,000,
and a voltage swing of only 1 V
pp
across the modulator achieves 6 dB on-to-off ratios at a 3 dB
insertion loss for non-return-to-zero (NRZ) binary data. The low voltage, near-zero quiescent
current, and low capacitance (15 fF, including wiring capacitance) result in an energy-efficient
modulator driven by a standard CMOS logic inverter at gigabit datarates using the same 1 V
nominal supply that powers digital electronics.
As a resonant device, the modulator is highly sensitive to c-Si layer thickness variations within and
across SOI wafers
24
as well as spatially and rapidly temporally varying thermal environments
created by the electrical components on the chip
25, 26
. Both effects cause λ
0
to deviate from the
design value, necessitating tuning circuitry. We embedded a 400 resistive microheater inside the
ring to efficiently tune λ
0
and added a monitoring PD weakly coupled to the modulator drop port.
When light resonates in the modulator ring, a small fraction of the light couples to and illuminates
the PD. This generates photocurrent proportional to the amount of resonating light, which is

Citations
More filters
Journal ArticleDOI
01 Jul 2017
TL;DR: A new architecture for a fully optical neural network is demonstrated that enables a computational speed enhancement of at least two orders of magnitude and three order of magnitude in power efficiency over state-of-the-art electronics.
Abstract: Artificial Neural Networks have dramatically improved performance for many machine learning tasks. We demonstrate a new architecture for a fully optical neural network that enables a computational speed enhancement of at least two orders of magnitude and three orders of magnitude in power efficiency over state-of-the-art electronics.

1,955 citations


Cites background from "Single-chip microprocessor that com..."

  • ...However, recent dramatic improvements in optical interconnects using integrated photonics technology has the potential to significantly reduce data-movement energy cost [46]....

    [...]

Journal ArticleDOI
24 Sep 2018-Nature
TL;DR: Monolithically integrated lithium niobate electro-optic modulators that feature a CMOS-compatible drive voltage, support data rates up to 210 gigabits per second and show an on-chip optical loss of less than 0.5 decibels are demonstrated.
Abstract: Electro-optic modulators translate high-speed electronic signals into the optical domain and are critical components in modern telecommunication networks1,2 and microwave-photonic systems3,4. They are also expected to be building blocks for emerging applications such as quantum photonics5,6 and non-reciprocal optics7,8. All of these applications require chip-scale electro-optic modulators that operate at voltages compatible with complementary metal–oxide–semiconductor (CMOS) technology, have ultra-high electro-optic bandwidths and feature very low optical losses. Integrated modulator platforms based on materials such as silicon, indium phosphide or polymers have not yet been able to meet these requirements simultaneously because of the intrinsic limitations of the materials used. On the other hand, lithium niobate electro-optic modulators, the workhorse of the optoelectronic industry for decades9, have been challenging to integrate on-chip because of difficulties in microstructuring lithium niobate. The current generation of lithium niobate modulators are bulky, expensive, limited in bandwidth and require high drive voltages, and thus are unable to reach the full potential of the material. Here we overcome these limitations and demonstrate monolithically integrated lithium niobate electro-optic modulators that feature a CMOS-compatible drive voltage, support data rates up to 210 gigabits per second and show an on-chip optical loss of less than 0.5 decibels. We achieve this by engineering the microwave and photonic circuits to achieve high electro-optical efficiencies, ultra-low optical losses and group-velocity matching simultaneously. Our scalable modulator devices could provide cost-effective, low-power and ultra-high-speed solutions for next-generation optical communication networks and microwave photonic systems. Furthermore, our approach could lead to large-scale ultra-low-loss photonic circuits that are reconfigurable on a picosecond timescale, enabling a wide range of quantum and classical applications5,10,11 including feed-forward photonic quantum computation. Chip-scale lithium niobate electro-optic modulators that rapidly convert electrical to optical signals and use CMOS-compatible voltages could prove useful in optical communication networks, microwave photonic systems and photonic computation.

1,358 citations

Journal ArticleDOI
08 May 2019-Nature
TL;DR: An optical version of a brain-inspired neurosynaptic system, using wavelength division multiplexing techniques, is presented that is capable of supervised and unsupervised learning.
Abstract: Software implementations of brain-inspired computing underlie many important computational tasks, from image processing to speech recognition, artificial intelligence and deep learning applications. Yet, unlike real neural tissue, traditional computing architectures physically separate the core computing functions of memory and processing, making fast, efficient and low-energy computing difficult to achieve. To overcome such limitations, an attractive alternative is to design hardware that mimics neurons and synapses. Such hardware, when connected in networks or neuromorphic systems, processes information in a way more analogous to brains. Here we present an all-optical version of such a neurosynaptic system, capable of supervised and unsupervised learning. We exploit wavelength division multiplexing techniques to implement a scalable circuit architecture for photonic neural networks, successfully demonstrating pattern recognition directly in the optical domain. Such photonic neurosynaptic networks promise access to the high speed and high bandwidth inherent to optical systems, thus enabling the direct processing of optical telecommunication and visual data. An optical version of a brain-inspired neurosynaptic system, using wavelength division multiplexing techniques, is presented that is capable of supervised and unsupervised learning.

862 citations

Journal ArticleDOI
18 Apr 2018-Nature
TL;DR: A way of integrating photonics with silicon nanoelectronics is described, using polycrystalline silicon on glass islands alongside transistors on bulk silicon complementary metal–oxide–semiconductor chips to address the demand for high-bandwidth optical interconnects in data centres and high-performance computing.
Abstract: Electronic and photonic technologies have transformed our lives-from computing and mobile devices, to information technology and the internet. Our future demands in these fields require innovation in each technology separately, but also depend on our ability to harness their complementary physics through integrated solutions1,2. This goal is hindered by the fact that most silicon nanotechnologies-which enable our processors, computer memory, communications chips and image sensors-rely on bulk silicon substrates, a cost-effective solution with an abundant supply chain, but with substantial limitations for the integration of photonic functions. Here we introduce photonics into bulk silicon complementary metal-oxide-semiconductor (CMOS) chips using a layer of polycrystalline silicon deposited on silicon oxide (glass) islands fabricated alongside transistors. We use this single deposited layer to realize optical waveguides and resonators, high-speed optical modulators and sensitive avalanche photodetectors. We integrated this photonic platform with a 65-nanometre-transistor bulk CMOS process technology inside a 300-millimetre-diameter-wafer microelectronics foundry. We then implemented integrated high-speed optical transceivers in this platform that operate at ten gigabits per second, composed of millions of transistors, and arrayed on a single optical bus for wavelength division multiplexing, to address the demand for high-bandwidth optical interconnects in data centres and high-performance computing3,4. By decoupling the formation of photonic devices from that of transistors, this integration approach can achieve many of the goals of multi-chip solutions 5 , but with the performance, complexity and scalability of 'systems on a chip'1,6-8. As transistors smaller than ten nanometres across become commercially available 9 , and as new nanotechnologies emerge10,11, this approach could provide a way to integrate photonics with state-of-the-art nanoelectronics.

630 citations


Cites background from "Single-chip microprocessor that com..."

  • ...The numbers indicate major fabrication steps in the order appearing in the process: (1) and (2), transistor and photonic isolation fabrication; (3) transistor frontend fabrication up to source/drain implant, including gate definition; (4) deposition, annealing and polishing of photonic polysilicon film; (5) polysilicon full and partial etching for forming strip and ridge photonic structures; (6) doping implants (P and N) for active photonics; (7) high doping implants (P++ and N++) and salicidation for both electronic and photonic devices; and (8) metallization....

    [...]

Journal ArticleDOI
30 Aug 2018-Nature
TL;DR: How optical metamaterials are expected to enhance the performance of the next generation of integrated photonic devices is reviewed, and some of the challenges encountered in the transition from concept demonstration to viable technology are explored.
Abstract: In the late nineteenth century, Heinrich Hertz demonstrated that the electromagnetic properties of materials are intimately related to their structure at the subwavelength scale by using wire grids with centimetre spacing to manipulate metre-long radio waves. More recently, the availability of nanometre-scale fabrication techniques has inspired scientists to investigate subwavelength-structured metamaterials with engineered optical properties at much shorter wavelengths, in the infrared and visible regions of the spectrum. Here we review how optical metamaterials are expected to enhance the performance of the next generation of integrated photonic devices, and explore some of the challenges encountered in the transition from concept demonstration to viable technology.

585 citations

References
More filters
Journal ArticleDOI
TL;DR: In this article, a numerical Kramers-Kronig analysis is used to predict the refractive index perturbations produced in crystalline silicon by applied electric fields or by charge carriers.
Abstract: A numerical Kramers-Kronig analysis is used to predict the refractive-index perturbations produced in crystalline silicon by applied electric fields or by charge carriers. Results are obtained over the 1.0-2.0 \mu m optical wavelength range. The analysis makes use of experimental electroabsorption spectra and impurity-doping spectra taken from the literature. For electrorefraction at the indirect gap, we find \Delta n = 1.3 \times 10^{5} at \lambda = 1.07 \mu m when E = 10^{5} V/cm, while the Kerr effect gives \Delta n = 10^{-6} at that field strength. The charge-carrier effects are larger, and a depletion or injection of 1018carriers/cm3produces an index change of \pm1.5 \times 10^{-3} at \lambda = 1.3 \mu m.

2,502 citations

Journal ArticleDOI
19 May 2005-Nature
TL;DR: Electro-optic modulators are one of the most critical components in optoelectronic integration, and decreasing their size may enable novel chip architectures, and here a high-speed electro-optical modulator in compact silicon structures is experimentally demonstrated.
Abstract: Metal interconnections are expected to become the limiting factor for the performance of electronic systems as transistors continue to shrink in size. Replacing them by optical interconnections, at different levels ranging from rack-to-rack down to chip-to-chip and intra-chip interconnections, could provide the low power dissipation, low latencies and high bandwidths that are needed. The implementation of optical interconnections relies on the development of micro-optical devices that are integrated with the microelectronics on chips. Recent demonstrations of silicon low-loss waveguides, light emitters, amplifiers and lasers approach this goal, but a small silicon electro-optic modulator with a size small enough for chip-scale integration has not yet been demonstrated. Here we experimentally demonstrate a high-speed electro-optical modulator in compact silicon structures. The modulator is based on a resonant light-confining structure that enhances the sensitivity of light to small changes in refractive index of the silicon and also enables high-speed operation. The modulator is 12 micrometres in diameter, three orders of magnitude smaller than previously demonstrated. Electro-optic modulators are one of the most critical components in optoelectronic integration, and decreasing their size may enable novel chip architectures.

2,336 citations

Journal ArticleDOI
01 Jun 2000
TL;DR: Optical interconnects to silicon CMOS chips are discussed in this paper, where various arguments for introducing optical interconnections to silicon chips are summarized, and the challenges for optical, optoelectronic, and integration technologies are discussed.
Abstract: The various arguments for introducing optical interconnections to silicon CMOS chips are summarized, and the challenges for optical, optoelectronic, and integration technologies are discussed. Optics could solve many physical problems of interconnects, including precise clock distribution, system synchronization (allowing larger synchronous zones, both on-chip and between chips), bandwidth and density of long interconnections, and reduction of power dissipation. Optics may relieve a broad range of design problems, such as crosstalk, voltage isolation, wave reflection, impedence matching, and pin inductance. It may allow continued scaling of existing architectures and enable novel highly interconnected or high-bandwidth architectures. No physical breakthrough is required to implement dense optical interconnects to silicon chips, though substantial technological work remains. Cost is a significant barrier to practical introduction, though revolutionary approaches exist that might achieve economies of scale. An Appendix analyzes scaling of on-chop global electrical interconnects, including line inductance and the skin effect, both of which impose significant additional constraints on future interconnects.

1,233 citations


"Single-chip microprocessor that com..." refers background in this paper

  • ...(2) The current detector is absorption-length limited19 and resonating the detector can improve sensitivity without an increase in the device size....

    [...]

Journal ArticleDOI
01 Jul 1984
TL;DR: The combination of decreasing feature sizes and increasing chip sizes is leading to a communication crisis in the area of VLSI circuits and systems, and the possibility of applying optical and electrooptical technologies to such interconnection problems is investigated.
Abstract: The combination of decreasing feature sizes and increasing chip sizes is leading to a communication crisis in the area of VLSI circuits and systems. It is anticipated that the speeds of MOS circuits will soon be limited by interconnection delays, rather than gate delays. This paper investigates the possibility of applying optical and electrooptical technologies to such interconnection problems. The origins of the communication crisis are discussed. Those aspects of electrooptic technology that are applicable to the generation, routing, and detection of light at the level of chips and boards are reviewed. Algorithmic implications of interconnections are discussed, with emphasis on the definition of a hierarchy of interconnection problems from the signal-processing area having an increasing level of complexity. One potential application of optical interconnections is to the problem of clock distribution, for which a single signal must be routed to many parts of a chip or board. More complex is the problem of supplying data interconnections via optical technology. Areas in need of future research are identified.

1,187 citations


"Single-chip microprocessor that com..." refers methods in this paper

  • ...(1) The current modulator design uses a mid-level p-implant (1017–1018 cm−3) for p-contacts as opposed to a p+ implant, creating high series contact resistance...

    [...]

Journal ArticleDOI
TL;DR: Results confirm the unique benefits for future generations of CMPs that can be achieved by bringing optics into the chip in the form of photonic NoCs, as well as a comparative power analysis of a photonic versus an electronic NoC.
Abstract: The design and performance of next-generation chip multiprocessors (CMPs) will be bound by the limited amount of power that can be dissipated on a single die We present photonic networks-on-chip (NoC) as a solution to reduce the impact of intra-chip and off-chip communication on the overall power budget A photonic interconnection network can deliver higher bandwidth and lower latencies with significantly lower power dissipation We explain why on-chip photonic communication has recently become a feasible opportunity and explore the challenges that need to be addressed to realize its implementation We introduce a novel hybrid micro-architecture for NoCs combining a broadband photonic circuit-switched network with an electronic overlay packet-switched control network We address the critical design issues including: topology, routing algorithms, deadlock avoidance, and path-setup/tear-down procedures We present experimental results obtained with POINTS, an event-driven simulator specifically developed to analyze the proposed idea, as well as a comparative power analysis of a photonic versus an electronic NoC Overall, these results confirm the unique benefits for future generations of CMPs that can be achieved by bringing optics into the chip in the form of photonic NoCs

873 citations