Shu Namiki - The Institute of Optics

Photonics technologies
for low-energy networks at AIST
Shu Namiki
National Institute of Advanced Industrial Science and Technology (AIST)
Network Photonics Research Center
ISUPT 2013, Oct. 21, 2013
At University of Rochester
Part of this work is supported by
p
Coordination Funds for Promoting
g Science and Technology
gy of Ministry
y of
- Special
Education, Culture, Sports, Science and Technology (MEXT).
- New Energy and Industrial Technology Development Organization (NEDO)
1
Team
•
Network Architecture
– Kiyo Ishii
– Junya Kurumida
– Tomohiro Kudoh
•
Silicon Photonics and Devices
–
–
–
–
–
–
–
–
•
Transmission and Node Subsystems
–
–
–
–
–
2
Hitoshi Kawashima
Keijiro Suzuki
G. W. Cong
Ken Tanizawa
Sang-Hun Kim
Haruhiko Kuwatsuka
Hiroyuki Matsuura
Aaron Albores-Mejia
Takayuki Kurosu
Takashi Inoue
Hung Nguyen Tan
Karen Solis-Trapala
Mingyi Gao
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Outline
• Introduction:
– Scalability, and now is endangered!
• Opportunity for optical switches
– Distinct energy scaling and bandwidth
– AIST project called “VICTORIES”
– Development of 32 x 32 silicon switches
• D
Dynamic
i O
Optical
ti l P
Path
th N
Network
t
k
• Inter-Data Center Network
• Summary
3
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Network traffic keeps on increasing
But, how sustainable is it for the future?
Revenue ($millions)
450,000 40% CAGR
400,000 AT&T
350,000 Verizon
300,000 Traffic (A.U.)
250 000
250,000 Gap
200,000 150,000 100,000 50,000 0 2007
2008
2009
2010
2011
2012
2013
Year
Source: Fortune 500
After T.J.Xia (Verizon) at WIN 2012, Inuyama
4
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Zettab
bytes/Yeaar
Data-centricity of Traffic
8
7
6
5
4
3
2
1
0
within DC
D2D (such as VPN)
Grobal IP DC
Grobal IP non‐DC
2011
2012
2013 2014
Year
2015
2016
And, most of the contents are video.
http://www.cisco.com/en/US/netsol/ns827/networking_solutions_sub_solution.html
http://www.cisco.com/en/US/netsol/ns1175/networking_solutions_sub_solution.html
5
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
How have we scaled the transport capacity?
• 1960s
– Copper wire for telephony
• 1970s
– Use optical fibers, LDs, PDs
• 1980s
– Use ETDM (Sonet/SDH)
• 1990s
– Use EDFAs, Internet revolution!!!
• 2000s
– Use WDM, Raman amplifiers, FEC, the Bubble!!!
• 2010s
– Use advanced modulation formats
formats, digital coherent technology
• 2020s
– Use Space and/or Mode-Division-Multiplexing
• We’ve been so acclimated to scalable technology, but…
6
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
What we are learning from HPC
• Because chips are no longer as scalable, Moore’s law
may finally be ending unless the network scales.
A. Benner (IBM)
A
(IBM), OFC 2012
Tutorial
http://www.extremetech.com/computing/165331-intels-formerchief-architect-moores-law-will-be-dead-within-a-decade
7
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Can data centers scale out?
• The role of network gets more important.
OIDA Workshop Report April 2013
8
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Fundamental limits being faced!
•
Quantum Limit
– Distributed Raman amplifiers with 3.1 dB noise figure
– E-FEC allows error free transmission with only several photons per bit
at receiver
•
Rayleigh Scattering Limit of Fiber
– ~0.15 dB/km
•
Shannon Limit
– A few dB away from E-FEC technologies
– Non-linear Shannon limit is about 100 Tbps/fiber
•
El
Electrical
i lB
Bandwidth
d id h Li
Limits
i = Th
The Li
Limit
i off P
Parallelism
ll li
– Radiation loss
– Thermal jitter
– Bandwidth density = energy consumption / bit
•
TCP/IP Limits
– Increasing inefficiency for larger Delay-Bandwidth products
– QoS limitations
– IP routers consumes energy proportional to the amount of traffic
9
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
• Disaster p
proof
– Communications is the
‘Lifeline’.
– Energy supply will be
limited when disasters
occur.
10
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
100,000
10,000
100,000
Japan’s Case
10,000
1,000
100
1,000
Japan’s Total
Electricity Generation
In 2005
Mostly HighDefinition Video
Transmissions
10
1.0
100
10
1
0.1
2000
2010
2020
2030
2040
0.1
2050
Total Internet Traffic （Tbps)
• Green of ICT, then
Green by ICT
Annuall energy consumpttion of IP Routers
（TWh）
Energy issue will eventually take over everything
• The current technologies can’t scale to the increasing traffic in future.
• 3-4 digit energy saving is necessary, which means we need a new paradigm.
Almost all of the traffic will be due to video transfer
Traffic Growth Projected by CISCO
Internet Trafffic (EB//Month)
1.5
Consumer Internet Traffic
W eb, Email, and Data
File Sharing(P2P)
Video
Gaming
Voice
1.0
Percentage
from 2008 to 2013
Video
CAGR=1.70
CAGR=1.39
0.5
Video
After MIC
(0.32 EB)
File Exchange
CAGR=1.21
Web, CAGR=1.24
0
2008
22.8%→61.6%
50.8%→24.7%
50 8%→24 7%
14.9%→8.2%
3.7%→3.7%
7 8%→1 8%
7.8%→1.8%
2009
2010
2011
2012
2013
source: Cisco Visual Networking Index: Forecast and Methodology, 2008-2013
11
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Packet switching versus circuit switching
IP: Process every
yp
packet for routing
g
Which suits better for video?
Optical Path: Set up end-to-end paths
12
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Switches: photonics vs. electronics
O ti l switches
Optical
it h
El t i l Switches
Electrical
S it h
Type of switching
Circuit switching
Packet switching
(requires a control plane)
Granularity
Coarse and inflexible
Fine and flexible
Bandwidth scaling
Transparent
Transceivers and LSI
Physical I/F
Optical connectors
Transceivers
Energy scaling
< W/port (incl. amp)
0.3 nJ/bit (Infiniband)
6 nJ/bit (IP router)
Energy
gy consumption
p
for 1,000 ports x 1Tbps
< 1 kW
300 kW ((Infiniband))
6,000 kW (IP router)
Scalability issue
Number of ports
Moore’s law
QoS
Physically guaranteed
Depends
Suitable apps
Video
Packetized data
Optical switch can save energy by several orders of magnitude!!
The hybrid use of packet and optical switching will be the key.
13
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Difference at the network level..
•
C
Conventionally
ti
ll exchange
h
digital
di it l packets:
k t Aggregate
A
t and
d Transport
T
t
A few lines between Routers
10
10-100
100 GbE
Statistical
Multiplexing
•
Bottle
Bottleneck
Statistical
Multiplexing
100-1000
100 1000 GbE
Bottleneck
Many CRS-3
required!!
h
E
h
dt
d numerous fibers
fib
i parallel!!
ll l!!
→AN
New S
Scheme:
Exchange
end-to-end
in
N users
Transmission lines
in parallel space
・ SDM rather than TDM
・ Full mesh connections
rather than
Aggregate and transport
m paths
Nxm
Matrix Switch
14
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Optical switches:
Transparent and Green
S. Namiki et al
S
al., ECOC 2010,
2010 Mo
Mo.2.A.4.
2A4
S. Namiki et al., JSTQE, 17, 446 (2011).
It calls for new network technologies
Energy and Cost
Efficiency
L3 (IP) Routing
L2 (Ethernet) Switching
Agility
Resilience
Good
Poor
L1 (ODU) Switching
Dynamic
Wavelength Path Switching
Optical
p
Path
Wave Band Path Switching
Network
Optical Path Switching
(DOPN)
15
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Good
Good Control
Plane at
upper
ppPoor
layers
y
&
Good Optical
Devices
Toward Optical Cut-Through of Everything (OCE)
• Today
Access
Edge
OLT
Splitter
p
ONU
16
Aggr
.
Aggr.
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Core
Access
Edge
OLT
Splitter
p
ONU
Toward Optical Cut-Through of Everything (OCE)
• Tomorrow
Access
Aggr
.
Aggr.
Edge
Core
Edge
OLT
OLT
Splitter
p
Splitter
p
ONU
17
Access
Optical Cut-through of Core Routers
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
ONU
Toward Optical Cut-Through of Everything (OCE)
• Ideally Energy-Saving DOPN
Access
Aggr
.
Aggr.
Edge
Core
Edge
OLT
DOPN-FE
OLT
DOPN-FE
Splitter
p
Splitter
p
ONU
Access
Optical Cut-through of Core + Edge Routers
Global Control Plane and Resource Management are necessary
necessary.
18
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
ONU
Jump from Optical Cut-through to DOPN
Core
Optical amplifier
138W/fiber/60km
・Avg. hop number is 4.2.
Transponder
42W/10Gbps
149W/40Gbps
351W/100Gbps(DC)
Core router
15.4nJ/bit
Model for Dynamic Optical Path Network (DOPN)
Regenerator
(comparable energy
consumption with TPND)
REG
Link distances: 150~850km
制御I/F
Big Datacenter
WSS
Data Center
Edge router
19.0nJ/bit
Metro-core
Optical Node
86W/degree
IE
Transponder
L2
ODUSW
*Considering low utilization of
down-link side ports
400W/equipment
A
Accommodating
d ti 410 ONU
ONUs
(80% utilization)
Logically, 4x4 regular lattice in core
with multi-granular switching.
OLT
Splitter
ONU
ONU
OLT
OLT
ONU
OLT
Splitter
Splitter
5W/equipment
Traffic distribution defined as a function
of population distribution
ONU
ONU
Access
Intelligent edge (IE)
Splitter
ONU
ONU
OpenFlowSW
OLT ODU Agg.
ONU
Path requests
Energy Consumpttion [TWh/year]
Level 2
Optical cut-through in core
Aggregation
OLT
200
ONU
150
Extrapolation of the
present model
100
50
0
2008
2010
2012
2014
2016
2018 2020
Year
2022
2024
2026
2028
2030
The energy reduction by optical cut-through is not large enough.
The energy consumption at edge routers is dominant.
19
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Energy Consumptio
on [TWh/year]
300
Level 1
Splitter
ONU
Calculation by K. Ishii (AIST),
Under the NEDO Green IT project.
j
Details to appear in ECOC2013.
250
IE
Small DC or
Content server
Aggregation
Edge switch 30nJ/bit*
ONU
IE
Splitter
ONU
IE
Calculation by K. Ishii (AIST),
Under the VICTOIRES project.
Details in Photonics West 2013.
ONU
Extrapolation of the present model
Optical cut‐through in core
Dynamic optical path network
1000
100
Yr 2030
Yr.
Yr. 2024
10
1
0
Traffic CAGR: 40%
10
20
Total Traffic [Pbps]
30
40
Some initiatives have already started
• GreenTouch Consortium
• VICTORIES
http://www.greentouch.org/
IEC’s
IEC
s Megatrend: Energy Sustainability of optical communication networks
and their driver: Dynamic Optical Path Network
http://www.aist-victories.org/
20
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Digital Optical Path Processor using Silicon Photonics
•
32 32 Matrix
32x32
M t i Switch
S it h
– Energy: Need only a few Ws
– Footprint: Within 2 cm x 1 cm Chip
•
Switching time
– ~ sec by thermo-optical switching
– ~ nsec by plasma-effect switching
•
Integration with control circuits
for black-box operations
Opitcal
Fiber Array
p
y
TO-Switch
Control Circuit
21
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Courtesy of Y. Shoji, K. Kintaka, et al.
Why 32 x 32 matrix switches?
•
•
•
32 x 32 SWs make a 512 x 512 Clos SW.
100,000 CPU nodes can be accommodated by 675 450x450
(simplex) switches
switches. A 450x450 SW consists of 96 32x32 SWs
SWs,
leading to a total of 64,800 32x32 SWs.
Each port can carry a bandwidth > Tbps.
2-stage Fat tree Switch Fabric for 100,000 nodes
450x450
450x450
450x450
225 switches
450(d)
450x450
duplex 225 ports
simplex 450 ports
22
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
450x450
450x450
450x450
450 switches
Total: 101250 CPU nodes
Roadmap of Si-photonics matrix switch at AIST
2014
Mach-Zehnder Interferometer
based TO switching
32×32
2013
2012
2011
2010
4×4
2×2
23
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
PILOSS
8x8
PILOSS
4×4
~20
20 mm
K. Suzuki et al., ECOC 2013, PDP
24
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
2×2 Polarization diversity
Kim et al., in ECOC2013
PS
2X2 Sw.Unit for TE
PS
TE
Intersection
PS
PS
2X2 Sw.Unit for TM
TM
Orthogonal SOP Crossing
（JP 2011
2011-158030
158030 ）
TM
25
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
TE
Si-ph. SW integrated with CMOS driver circuits
•
OXC: 2x2 TO switch with MOSFET
– G. W. Cong et al., OE March 2013 (AIST)
•
Interconnect SW: 4x4/8x8 p-i-n switch with CMOS driver
– B.G. Lee et al., OFC NFOEC PDP 5C3, 2013 (IBM)
26
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Realizing 32 x 32 chip scale
Drawing time using Spot-Beam EB
Lithography System (JEOL6000FS/E (50kV))
8×8
11 mm×7 mm
4×4
6 mm×3 mm
27
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
128 hrs. (～5 days) by EB
8 hrs.
EB
2 hrs.
32×32
20 mm×10 mm
DUV photolithography
ArF Immersion Lithography facility
Super Clean Room (SCR) 12 inch line ＠AIST
Hierarchical multi-granularity in routing
• Various path granularities
– From fine to coarse
－ node that can handle such
Control
I/F
制御I/F
Sub-lambda path
1GE
10GE
10GE
ODU
ODU
OTN
ODU
OTN

path

path
100Gbit/s~
Fiber
path
10~100 Gbit/s
WSS
Fiber
path
光パス
Lambda
path
波長パス
Optical SW
光スイッチ
Transponder
L2
ODU
SW
Lambda SW
（WSS）
Fiber path SW
（Matrix SW）
Electrical
SW
電気スイッチ
ODUSW
1~10 Gbit/s
サブ・ウェーブレングス・パス
Sub-lambda path
~1Gbit/s
MPLS-TP
K. Ishii et al., Photonics West 2013
• Network topology based on hierarchical routing
– Low
L
energy consumption
ti
28
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
－Scalability
S l bilit tto national
ti
l scale
l
CDC-ROADM using Silicon Photonics (NEC+AIST)
1:8 Splitter
Next
Gen. ROADM
次世代ROADM
8x1 Switches
10-5
…
…
WSS
/Splitter
WSS
/Splitter
BER
10-6
λ1~λ8
10-7
10-8
10-9
λ1~λ8
10-10
Out-Port 3
Out-Port 4
Out-Port 6
Out-Port 7
Out-Port 5
10-11
In-Port 1
In-Port 8
λ1
1:8 Splitter
…..
…..
48 transponders
48 transponders
16 mm
8x1 Switches
Add
Drop
…
…
…
TPA
TPA
TPA
TPA
TPA
TPA
module
module
module
module
module
…
…
module
…
…
…
..
..
..
..
…
..
To transponders
(8-port)
Splitter
1 x 8 CPL
…
12 mm
152 switch elements
integrated
1 x 8 CPL
Si-TPA transmission card
…
…
To input/output
fibers
(8-port)
λ8
43.018 Gbps DPSK signal
OSNR 21 dB @ res 2 nm
Received Power: ‐9dBm
..
Transponder
Gate / Selector
8 Input/output
fibers
SW
Drivers
48 transponders
TPA module x3 / daughter card
(8 x 8 Si switch module)
T. Hino et al., ECOC2012 Invited：Tu.3.A.5
29
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
TPA module x12
Switch drivers
FPGA
Feature of ODU XC (Fujitsu+AIST)
•
ODU XC (Cross Connect) is a digital cross
connect for ODU (Optical Data Unit) that defined
by ITU-T G.709.
•
ODU XC has capability of switching traffic with
1.25Gbps granularity. It is very efficient to switch
1Gbps-40Gbps stream.
•
ODU XC has 1/3 power consumption and 1/100
latency compared to core router.
OTN
λ
Path
Optical
Fibre
ODU XC
ODUk(k=0,1,2,3,4,flex)
ODUk(k=0,1,2,3,4,flex)
ODUk(k=0,1,2,3,4,flex)
ODUk(k=0,1,2,3,4,flex)
ODUk(k=0,1,2,3,4,flex)
ODUk(k=0,1,2,3,4,flex)
ODUk(k=0,1,2,3,4,flex)
ODUk(k=0,1,2,3,4,flex)
…
10GE
ODU2
ODU
Packet#1 ODUflex#1 XC XC
Router
R
t
100GE
Packet#2 ODUflex#2 node
Router
λ
Path
λ
ODU ODUflex#1 Packet#1
ODUflex#3 Packet#3
XC
XC node
10GE Packet#3 ODUflex#3 ODU
XC 40GE Packet#2 ODUflex#2 node
λ
ODU
XC node
40GE Router
λ
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
OTN
…
Slide courtesy of H. Honma
and H. Onaka, Fujitsu
30
10GbE
AIST and Fujitsu are developing the ODU XC chip
and this chip
p will be available in 2013 .
OTN is effective and efficient
f operating >1 GbE.
for
G
Fujitsu
plans to incorporate the OTN
switch fabric into Fujitsu
products.
ODU
ODU
ODU
λ
•
1GbE
1GbE
ODU2
10GE
Router
32 x 32 SWs make up a nationwide Dynamic
Optical Path Network
R l
Replace
IP routers
t
by
b optical
ti l switches!!
it h !!
Hierarchical multi-granular node
Sub- aggregating node
Sub- and wavelength cross-connect node
Wavelength path terminal
Multi-granular aggregating node
Sub- path terminal
Content server
…
Core
…
…
…
…
…
…
n
…
C-plane I/F
…
…
Fiber path
100Gbit/s~
WSS
…
10~100 Gbit/s
-path
Optical
Transponder
1
Category-core
2
3
Group 1
1 2
31
…
…
k
1
…
1 2
i
1
k
j
…
3
1
m
mx mx
+1
i 1 2
2
…
…
mx
i 1 2
2
…
…
1 2
i
3
1
m
mx
+1
Electrical
1~10 Gbit/s
k
…
…
j
…
mx
…
…
1
2
3
K. Ishii, et al.(AIST),
k
… Photonics West 2013
1 …
…
m
mx
+1
i 1 2
Sub- path
1
j
1
1
L2
…
3
j
1
Category
2
ODUSW
OLT
i
Passive
Double Star
Regular mesh in Core; Full Mesh in Metro; Complete bipartite graph in Access
Flex-Grid compatible, ODU-XC for sub- granularity
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Inter-datacenter network: challenges
•
•
•
•
•
Chunkk off continually
Ch
ti
ll generated
t dd
data
t
Ultra-coarse granularity
Low latency data transfer
High capacity and efficiency
Strict QoS with scheduled connections
Datacenters
Distributed
Datacenters
•
•
32
Seamless connections to intra
intra-datacenter
datacenter network
Resilience for disaster
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Multi-Tbps Elastic Transmitter : WDM vs. OTDM
• Super-channel
S
h
lN
Nyquist/OFDM
i /OFDM WDM
xN carriers
Nxm
Matrix
SW
Mod.
・・・
Comb
Gen.
・・・
AWG
Mod.
PDM
Combiner
Mod.
DSP hungry
• Nyquist-OTDM
Nyquist OTDM WDM
xN symbol slots
Mod.
1x
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
・・・
Splitter
33
0x
・・・
Comb
Gen.
Mod.
Waveshaper
+ PDM
Combiner
M d
Mod.
mx
Optical processing
needed?
Optical comb generator (OCG) for Tbps elastic Tx/Rx
•
•
High spectral efficiency through superchannel (OFDM or Nyquist)
High quality optical comb/ultra-short pulse through highly nonlinear fibers
– V. Ataite et al. (UCSD), OFC NFOEC 2013, PDP5C.1
C+L-band
40 dB OSNR
– T.
T Inoue and S.
S Namiki,
Namiki Laser & Photon
Photon. Rev.,
Rev 2: 83
83–99
99 (2008)
(2008).
< 400 fsec
Tunable
•
34
R&D on Tbps elastic transceivers using OCG is highly expected.
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Research on Nyquist OTDM‐WDM signals
Features of Nyquist OTDM
o High spectral efficiency up to 1 symbol/s/Hz
o Ultrahigh yet flexible baudrates (up to 1T)
o Generation in optics, high energy efficiency
o Eliminating guard‐bands in WDM system
o Better transmission perf. than OTDM
o T. Hirooka
T Hi k et al., Opt. Express, 20, 2012.
t l O t E
20 2012
o K. Harako et al., OFC 2013, JW2A.38.
o H. Hu et. al., CLEO 2013, CTh5D.5.
M. Nakazawa et al., Opt. Express, 20, 2012.
G. Bosco et al., J. Lightw. Technol., 29, 2011.
Nyquist OTDM‐WDM in network
i
i
k
o Granularity will grow to be more than ~100G in the future big‐data centric era
o Especially, traffic between data‐centers
Especially traffic between data‐centers
o Highly efficient, ultra‐coarse yet flexible granular inter‐datacenter network based on all‐optical elastic network
 Nyquist OTDM‐WDM signals
35
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
O. Gerstel, et al., Comm. Magazine, 50, 2012.
Nyquist filtering using Waveshaper
Ultra‐coarse granular
Nyquist signals
yq
g
o Ultra‐coarse granular Nyquist signals (>40G) can be generated by Waveshaper
Ultra coarse granular Nyquist signals (>40G) can be generated by Waveshaper
o Waveshaper‐based WSS can efficiently add‐drop WDM channels of Nyquist OTDMs with almost no guard‐band
o Pass‐drop over cascaded WSSs: H. Nguyen Tan et al., OFC 2013, JW2A.50
Pass drop over cascaded WSSs: H Nguyen Tan et al OFC 2013 JW2A 50
 This paper: transmission and pass‐drop operations in full elasticity
36
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Experimental setup
o Baudrate/pol. : 43Gbaud, 86Gbaud, 172Gbaud, and 344Gbaud
o Clock recovery by optical null‐header (ONH): Kurosu et al., Opt. Express, 21, 2013.
o Transmission 4 spans of 80km SLA+IDF and WSS nodes
Transmission 4 spans of 80km SLA+IDF and WSS nodes
o WSSs elastically pass‐drop WDM channels of Nyquist OTDMs
37
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Optical spectra of Nyquist OTDM WDM
H. Nguyen Tan et al., OECC 2013, PDP / ECOC We.1.C.5
Details to appear in OE.
INPUT
WSS node
PASS
DROP
o Simple generation of different baudrate signals by Waveshaper
o WSS efficiently add‐drop WDM channels of Nyquist OTDMs with almost no guard‐band
38
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
BER performance of Nyquist OTDMs
ONH
ONH
ONH
39
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
VICTORIES Test Bed DEMO2014
• Oct.
O t 2014 iin T
Tsukuba,
k b JJapan
• Co-located with a 3-day international symposium
• Demonstrate 4k/8k video apps over DOPN
ネットワークリソースマネージメント
I1：ストレージ資源管理（AIST）
N2：ダイナミック波長パス
N2：ダイナミック波長パス・
ノード技術（NEC）
ITRI M i Direction: 3
ITRI‐Main
Di ti 3
ITRI‐Main‐TKB
‐WEST(NS1)
Agg.
I2:ストレージ・ネットワーク統合資源管理（NTT）
TL‐1
N1：多粒度ダイナ TL‐1
ミック・ノード（AIST）
P5：集積デバイス実装
（古河電工）
NPRC‐Main
Agg.
P3：トラポン・アグリゲータ（NEC）
Transponder
ODU SW
TL‐1
P1, P2：シリコンフォトニクス
TOスイッチ（AIST）（富士通研）
Transponder
N4：ダイナミック波長可変光源（住友電工）
ODU SW
NPRC EAST1
NPRC‐EAST1
Direction: 3
Agg.
P4：次世代
WSS
（日立電線）
N3：サブウェーブレングス
・パス・ノード（富士通）
配信サーバA
(1G, HD)
配信サーバC
(10G, HD)
配信サーバB
(10G, HD)
視聴者SA (8K)
Direction 2
Direction: 2
TV会議SYS2
視聴者A
AIST情報棟
視聴者B
コンテンツ集積
C5：超高速波形観測技術（Alnair）
Mon.
拠点(HD)
Direction: 3
サーバS
ITRI‐WEST
NPRC‐CTR
Agg.
Agg.
ODU SW
C4：集積型分散モニタ技術（フジクラ）
C3：特殊ファイバ、高密度
線路技術（古河電工） NPRC‐EAST2
Direction: 4
Direction: 3
Agg.
MUX
C6：光パス多重
技術（富士通研）
視聴者C
C1：自律制御光パス・
コンディショナ開発（AIST）
視聴者SB (8K) 視聴者E
C2：高速制御
技術（Trimatiz）
AIST デ
デモ会場
会場
視聴者D
TKB‐SUB1
TKB‐SHV
Direction: 3
TKB‐SUB2 Direction: 2
Direction: 2
Agg.
Agg.
TV会議SYS1
配信サーバ
(8K, SHV)
40
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
視聴者SC (8K)
・制御要素の明確化（波長，帯域，ポート＃，
制御 素
確 波
帯域 ポ
フォーマット，ビットレート，ファイバ種，分散量）
視聴者F
つくば地区
協力機関
Summary
•
•
•
•
•
•
•
•
•
41
The energy consumption of the IP network is becoming a serious bottleneck,
calling for a paradigm shift.
For disaster survivability, resilience under limited energy supply is of critical
importance.
importance
Dynamic Optical Path Network may be the only option in the sense of both
energy, capacity, and suitability for video services, thus has been recognized
g
byy IEC as a driver of this megatrend.
In order to realize this, some R&D initiatives have been put into action.
We have reviewed some of the activities under VICTORIES project at AIST.
An extensive effort on developing 32 x 32 silicon photonics switches was
reviewed.
All optical Nyquist filtering works well higher bandwidths than 40 Gbaud
signals.
Jump from ongoing optical cut-through to Dynamic Optical Path Network must
be assisted through academic activities and standardization!
Orchestration of different layers and/or standardization bodies is essential.
S. Namiki, ISUPT2013, Oct. 27, Univ. of Rochester
Thank you!
[email protected]
42
S.
Namiki, ISUPT2013, Oct. 27, Univ. of Rochester