Upload
others
View
9
Download
0
Embed Size (px)
Citation preview
Huawei Optical IntelligenceWhite Paper
Contents
Trends and Challenges
Huawei Optical Intelligence Solution
Premium Private Lines3.1 Latency Map: Improving Network Monetization Capability
3.2 CPE Plug & Play: Reducing Site Visit of Software Commissioning Personnel and Accelerating
Service Provisioning
3.3 Success Story: Premium Private Line Increases Carriers’ Revenue and Profit
4.1 Optical Network Health Assurance: from reactive to Predictive O&M, Reducing OPEX and
Improving Customer Experience
4.2 Resource Assurance, Automatic Resource Bottleneck Discovery, and Precise Capacity
Expansion
4.3 Highly Reliable Network, Improving Service Reliability
4.4 Success Stories
4.4.1 Huawei and an European Tier-1 Carrier Successfully Completed the First AI-based
Intelligent O&M and Joint Test for an Optical Network
4.4.2 Highly Reliable Network Solution with Reliability Improved from 98.5% to 99.99%,
Saving Annual Cost by 40M+
Zero-Touch Maintenance
Prospects
01 02
02
03
04
05
04
15
07
08
09
10
11
12
13
14
14
14
15
01
ContentsHuawei Optical IntelligenceWhite Paper
02
Trends and Challenges01With the rapid development of Internet+, 5G, 4K, and VR services, OTT players have started to tap into
the telecom market. Rapid service rollout in Internet mode and fierce competition brought by extensive
service innovations drive carriers to perform Internet-oriented transformation. Meanwhile, carriers face
the dilemma of sluggish revenue growth in traditional telecom services and constantly high O&M costs.
Both external and internal factors post many challenges to carriers:
On one hand, network investment keeps growing, innovation of new services is weak, and service
provisioning and innovation rely heavily on a few comprehensive vendors. "Selling whatever services
you have" is a typical "push" mode, which cannot keep up with the development of the Internet or
meet user demand. In the Internet+ era, service requirements are ever-changing and new service
opportunities are fleeting. Fast and agile service provisioning is the cornerstone of rapid service rollout.
Therefore, service provisioning efficiency must be measured by days and minutes. For example, for
latency-sensitive financial services, network latency must be accurate to sub-microseconds to ensure
core competitiveness in financial markets.
On the other hand, the network, as the ultimate entity that carries bandwidth traffic, becomes more
and more complex: The network topology is evolving from chain and ring to mesh and 3D-mesh. The
increase in network complexity multiplies O&M costs. As network O&M has exceeded the reasonable
capacity of "manual processing", carriers urgently need automation measures to reduce the skill
requirements for O&M personnel and effectively cut OPEX in a long run.
Figure 1-1 Trends and challenges of optical network development
Resource
From "chain" to "mesh" architecture
How to ensure that network resources
are ready?
Service
From“dumb pipe”to
"premium experience"
How to shorten service TTM and
improve user experience?
From manual to AI-powered/automated O&M
From reactive to proactive O&M
Maintenance
70% of failures are caused by human errors.
90% of maintenance are passive.
How to improve maintenance
efficiency and reduce costs?
Optical Intelligence (OI)
Trends and ChallengesHuawei Optical Intelligence
White Paper
03
According to the IHS 2018 survey report, optical network automation is increasingly valued by carriers.
The survey results are as follows:
50% of carriers believe that the main driving forces for the deployment of optical network
automation are as follows:
- Higher network O&M efficiency
- Service provisioning automation
- Network capacity planning automation
- Quicker introduction of new services
Source: IHS market survey, 2018
80%+ of respondents believe that the following automation features will be deployed around 2020:
- Real-time network resource visualization
- Real-time network planning
- Network health prediction and proactive O&M
- Enhanced, dynamic network restoration and centralized path computation
Figure 1-2 Driving forces for the deployment of optical network automation
76%
71%
67%
52%
43%
33%
0% 20% 40% 60% 80%
Transport SDN Deployment Drivers
Operating our networks with greatercapital
and operational efficiency
Simplification and/or automation
of service provisioning
Simplification and/or automation
of network capacity planning
Quicker introduction of new services
for faster time to revenue
Coordination/orchestration of services
that span multiple layers
Coordination/orchestration of services that
span multiple network operators
Trends and ChallengesHuawei Optical IntelligenceWhite Paper
04
Huawei Optical Intelligence SolutionHuawei Optical Intelligence
White Paper
Figure 1-3 Time to deploy optical network automation
Source: IHS market survey, 2018
In a word, carriers require the next-generation optical network to empower intent-driven automation,
intelligence based on big data and Artificial Intelligence (AI), agile interconnections based on open and
programmable features, and high reliability based on dynamic restoration, achieving efficient O&M. An
optical network has been used only as a dumb pipe for a long time. Network deployment is
time-consuming and labor-consuming, service provisioning is slow, and fault locating is difficult.
Therefore, the software solution oriented to traditional static networks can no longer address network
evolution and service requirements. Continuous evolution is required for full-lifecycle automation.
Huawei Optical IntelligenceSolution02In response to carriers' challenges and requirements on networks and services during the
transformation process, Huawei has proposed the Optical Intelligence solution to promote optical
networks to focus on user experience, aimed at building an intent-driven, closed-loop system that
supports full-lifecycle end-to-end (E2E) automation. This solution helps carriers implement pipe
monetization, improve service experience, reduce network O&M costs, and maximize business value.
05
Figure 2-1 Evolution of the transmission network software architecture
Figure 2-2 Hierarchical architecture of Huawei Optical Intelligence solution
The software architecture of Huawei Optical Intelligence solution consists of the Network Cloud Engine,
device edge intelligence, and core technologies of AI algorithms, big data, and computing power (ABC)
to support full-lifecycle automation use cases, implement premium private lines and zero-touch
maintenance, helping carriers increase revenue and reduce O&M costs.
NMS
QxDCN
Data plane
NMS
QxDCN
Data plane Data plane
Device control plane Device control plane
ManagerController
Analyzer
Evolution Evolution
GMPLS/ASON control and
service self-healing
Integrated manager, controller and
analyzer, and full-lifecycle
automation and intelligence
enabled by AI and optical sensors
Static, manual management
Huawei Optical Intelligence SolutionHuawei Optical IntelligenceWhite Paper
Therefore, to stay relevant to ever-changing user demand, the transmission network software
architecture needs to evolve from static management to integration of manager, controller and
analyzer, and multi-layer intelligence, making optical networks increasingly automated and intelligent.
Manager Controller Analyzer
Optical sensors
ASON 2.0
Network Cloud EngineIndustry's first enhanced controller with
integrated management, control, and analysis.
Scenario-based app, enabling zero-touch
operations.
Intelligent physical networkASON 2.0, scalable large-network mgmt.,
speedy self-healing (optical < 10s, electrical <
200 ms), smart & reliable.
E2E network monitoring through leading
optical sensors.
Capabilities of ABC AI algorithms covering all scenarios of optical
networks.
Big data: 10000+ network-wide optical data
collected in every second.
Computing power 10 times higher than the
industry average.
Networkintelligence
Digitaltwin
Physical network
06
Figure 2-3 Full-lifecycle automation use cases of Huawei Optical Intelligence solution
Figure 2-4 Technical leadership of Huawei Optical Intelligence solution
The full-lifecycle automation use cases of Huawei Optical Intelligence solution cover two parts:
Premium private lines, which increase carrier revenue.
Zero-touch maintenance, which reduces OPEX through automation and intelligence.
Huawei Optical Intelligence solution leverages core technologies of ABC to enable full-lifecycle
automation and intelligence, achieving network autonomy in the end.
Use case
Leading AI
algorithms
Rich optical
big data
Excellent
computing
power
Optical network health prediction
Regression algorithm for
optical network
Faster data collection, more accurate reproduction of optical network analog signals.
Reinforcement learning and
optimization algorithm
Time series prediction algorithm
Aggregate algorithm for
optical network
Neural network algorithm for
optical network
Smart Commissioning Root cause analysis Resource Analysis
Optimal power spectrum model of deep neural networks
Evaluate Deliver
Feed back
Value
time
Collection withinseconds
15-minutecollection
Faster sampling: 15 min./time to 1s/time.More accurate : estimated to directly measured (such as OSNR and SOP).
Big data collection
Huawei Ascend 310 AI chips based acceleration boards provide 10+ times of computing power than industry.
Edge intelligence (equipment side)
500k+ online optical network devices deployed globally, rich big data for machine learning & modeling.
Inventory big dataOnlinedevices Data-lake
AI @ cloud
AI @ edge
Huawei Ascend 910, a single AI chip with the highest computing density is launched. (256T vs. Industry's best 125T)
Cloud intelligence (sever side)
Huawei Optical Intelligence SolutionHuawei Optical Intelligence
White Paper
07
Application: Huawei Optical Intelligence solution provides a variety of applications throughout the
entire process.
- Premium private lines: Pipe monetization is achieved through the latency map, CPE plug and play,
OVPN, and SLA assurance.
- Zero-touch maintenance: A bold new O&M mode is created to evolve from passive O&M to
predictive and proactive O&M, build an E2E full-lifecycle, automatic closed-loop process, and reduce
OPEX.
- Resource analysis and prediction: Capacity expansion is performed in advance to improve resource
utilization and shorten the TTM.
AI algorithms: As the key technologies that enable intelligence, the AI algorithms can extract
features from a large amount of data and create models from existing fault features or resource
features based on expert experience to quickly resolve network issues. In addition, the AI algorithms
based on machine learning can create a knowledge map and predict trends to prevent faults and
optimize performance in advance.
Big data: Physical feature data such as optical network parameters, optical power, OSNR, BER,
optical spectrum, SOP, and oDSP are collected in seconds through optical sensors on the physical
network to obtain real-time monitoring parameters. Based on technologies such as big data mining
and AI algorithms, the collected data is trained and models are created, which can be better used in
upper-layer applications to improve automation and lower OPEX.
Computing power: An increase in AI algorithm complexity and the processing of massive data
require high computing power. Huawei-developed AI chips provide powerful computing power for
big data collection, storage, analysis, training, and reporting.
Premium Private Lines03Currently, an optical network gradually evolves from a dumb pipe provider into an E2E OTN service
network, which helps carriers build premium OTN private lines for high-value customers such as
financial institutions, government agencies, and large enterprises. This enables carriers to implement
pipe monetization, maximize business value, and provide excellent user experience.
Figure 3-1 E2E schematic diagram of Huawei premium private lines
Premium Private LinesHuawei Optical IntelligenceWhite Paper
For financial services, network latency, one of the most demanding indicators for premium private lines,
must be accurate to sub-microseconds. Premium OTN private lines have the lowest latency. Huawei
premium private lines provide an E2E latency solution that can be sensed, sold, committed, and
guaranteed.
Figure 3-2 Latency of premium OTN private lines
Pre-sale
Q2O(quote to order)
Latency map Online portal OVPNOnline E2E resource
confirmation
In-sale
O2A (order to activation)
Automatic, E2E service provisioning
CPE plug and play Visualized provisioningprocess
After-sale
T2R (Trouble to
Resolution)
SLA assurance BOD Smart doctorLatency/Traffic/Status
3.1 Latency Map: Improving Network Monetization Capability
Latency Map
Real-time physical network latency detectionGoogle Maps-style latency map app
Latency Guaranteed
Real-time monitoring of service latency KPIsWarning and elimination of violation risks
Latency Routing Policy
Minimum-latency policyLatency-range policy (20–25 ms)
08
Premium Private LinesHuawei Optical Intelligence
White Paper
1.8 ms
Figure 3-3 Schematic diagram of CPE Plug & Play
09
Huawei Optical Intelligence solution provides the CPE plug-and-play capability to support rapid
provisioning of private line services. Hardware personnel need only to take CPEs to sites and install and
power on the CPEs. Devices at the central office (CO) automatically discover CPEs and CPEs
automatically go online through software protocols. In the end, software is automatically configured
and commissioned and services are automatically created to provide an IT-based, HBB-like, and
self-service user experience. For carriers, this solution reduces site visit costs of software commissioning
personnel, and shortens the service provisioning time.
3.2 CPE Plug & Play: Reducing Site Visit of Software Commissioning Personnel and Accelerating Service Provisioning
Complex CPE deployment, time-consuming service provisioning, and multiple site visits
RMS
Network management center
· Multiple rounds of communication· Manual data uploadManual data
input
CPE CPECO CO
Multiple site visits· Installation and NE ID configuration· NE commissioning and configuration· Meter-based acceptance test
Step Mode Time/CPE
Onsite CPE installation, power-on, and fiber connections
ID/IP address configuration and NE go-live
By software commissioning personnel
0.5 hour
By hardware installation personnel
1 person-day
CPE parameter configuration
Resource scheduling + service configuration
By software commissioning personnel
0.5 person-day
By software commissioning personnel
0.5 hour
Commissioning, meter-based testing, and acceptance for talent services
Customer service personnel + users
1 person-day
Site visit needed
As Is
3.5 days/person + site visits by software commissioning personnel
Step Mode Time/CPE
Onsite CPE installation, power-on, and fiber connections
ID/IP address configuration and NE go-live
Automatic configuration and go-live
Minutes
By hardware installation personnel
Hours
CPE parameter configuration
Resource scheduling + service configuration
Automatic configuration by software
Minutes
Automatic configuration by software
Minutes
Commissioning, meter-based testing, and acceptance for talent services
Online commissioning
Minutes
Free from site visits
Plug-and-play CPEs, check-free and configuration-free private line provisioning, and
meter-free test
To Be
Real-time resource visualization
Automatic NE ID allocation and NE go-live
ODUk pipe
Metro network A
Metro network Z
Backbone network
CPE CPECO CO
Finished in hours
Only one site visit· Onsite installation and power-on, zero-touch provisioning· Automatic configuration delivery· Meter-free Y.1564 test
Pre-occupancy of inter-CO pipe resources and completion of a line test
Premium Private LinesHuawei Optical IntelligenceWhite Paper
Figure 3-4 Product package design
Carrier U, a tier-1 carrier in China, is a pioneer in city B's private line market, which is the carrier's top
priority. Currently, the private line in city B is developing steadily. However, carrier U faces external and
internal challenges:
Price competition and product homogeneity with other carriers cut carrier U's profit.
Insufficient service innovation or flagship products cannot attract high-end users.
The provisioning of private line services is slow, decreasing user satisfaction.
In 2018, carrier U and Huawei joined hands to innovate premium private lines as follows:
Product package innovation: Basic bandwidth + Intelligent speed adjustment + Value-added services,
implementing differentiated product combinations to meet the appetite of high-end users. In
addition, the self-service mode goes through the pre-sales, sales, and after-sales phases to improve
purchase and usage experience of subscribers.
Feature competitiveness innovation: The latency map is a selling point of premium private line
services of carrier U, which provides ultra-low latency private lines for high-end financial customers.
Provisioning speed innovation: The E2E service configuration implements the automatic process
from the pipes to the services. In addition, the plug-and-play CPE reduces the site visits of software
commissioning personnel, saves the labor cost, and enables fast service rollout.
Business value:
The private line service provisioning time is shortened from weeks to two days.
The reliability of private line services is improved to 99.99%+.
In 2019, the revenue is expected to increase by USD 70 million, with a share increase of 12% and a
premium of 20%.
3.3 Success Story: Premium Private Line Increases Carriers’ Revenue and Profit
BandwidthLatency
predictionIntelligentspeed-up
Basic Basic Value-addedservice
Value-addedservice
Visualizedindicators
Value-addedservice
Securityencryption
Value-addedservice
Latencyoptimization
Smartdetection
Key guaranteed
service
10M
50M
100M
200M
500M
1G
10G
1 ms
2 ms
3 ms
Instantspeed-up
Bandwidthusage
L1encryption
1 ms
2 ms
3 ms
Productsubscription
Scheduledspeed-up
None
Visualizedlatency
Runningreport
Circuit status OVPN
Noencryption
√ √
√
√ √
√ √
√
√
10
Premium Private LinesHuawei Optical Intelligence
White Paper
Figure 3-5
Price comparison between existing private line products and new private line products
11
10M 20M 50M
Price
Unit: CNY
Private line level
Price comparison between existing private line products and new private line products
Existing private line products New private line products
Zero-Touch Maintenance04An optical network is a highly dynamic, time-evolving, complex analog system, which features complex
non-linear and coupling effects, as well as complex high-dimensional action space and status space and
involve thousands of states and parameters. Therefore, high engineer skills are required for network
O&M. Fault locating is also a systematic project, which is time-consuming.
Huawei Optical Intelligence features user-centric E2E full-lifecycle automation, enables predictive E2E
closed-loop O&M before fault occurrence and optimize quality before quality deterioration, minimizing
fault loss, and reducing OPEX.
Figure 4-1 Panorama of full-lifecycle automation of Huawei Optical Intelligence
Deployment
Provisioning
Monitoring
Guarantee
Analysis
Planning
User-centricResource visualization
Capacity prediction
Online planning
Resource automation
Automatic service provisioning
SLA assurance
Service automation
Health visualization
Health prediction
Automatic commissioning
Highly reliable network
O&M automation
Zero-Touch MaintenanceHuawei Optical IntelligenceWhite Paper
Figure 4-2 Optical network health prediction
The O&M of optical network have been reactive for a long time. Maintenance is performed only after a
fault occurs or a user registers a complaint. Carriers cannot identify the tell-tale signs and have to wait
for sub-healthy fibers or optical services to deteriorate until a fault occurs. A large amount of
compensation for breach of contract is often incurred from SLA violation, which increases the
maintenance cost. (According to the analysis of the network fault data of a carrier, it is found that the
fiber faults account for 68% of the total network faults. Gradual OTS/OCh faults account for 56% of all
fiber faults, and 38% of the total network faults. Among them, the bending, shaking, loosening, and
fiber core faults account for 90%.)
Huawei optical network health assurance package provides closed-loop OTS/OCh health monitoring,
sub-health prediction, and automatic optimization based on use cases such as visualized optical
network health, optical network health prediction, and intelligent optical network commissioning. The
optical network health prediction uses the machine learning and AI prediction algorithms to analyze the
health status of each optical fiber and channel. Based on the change trend of optical performance, the
system predicts the risk of faults and specifies faults points in advance. In this way, the system can
prevent network risks, provide recovery suggestions, implement proactive O&M, and reduce service
interruption, avoiding the compensation caused by SLA violation.
4.1 Optical Network Health Assurance: from reactive to Predictive O&M, Reducing OPEX and Improving Customer Experience
CurrentHistorical Forecast
Healthy
Faulty Warning
Actual value
Predicted value
Days/Weeks/Month
Sub-healthy
AI algorithm + ModelEnhanced logistic regression algorithm + Improved pattern recognition prediction algorithm based on time
series + …
Health model Sub-health model Fault model
Data cleansing and analysis
(Data parsing, format conversion, deduplication, and removal of invalid data)
Data production and collection (Object: OTS, OCh, optical amplifiers, WSS, OTU)
OTS: second-level optical power, OTDR fiber fault data, and millisecond-level SOP
OCh: second-level BER, OSNR, and alarm data (service interruption, low optical power, etc.)
12
Zero-Touch MaintenanceHuawei Optical Intelligence
White Paper
13
Figure 4-3 Resource assurance
With the rise of OTT players in the Internet industry, new challenges are posed to traditional carriers'
services. The TTM (1 to 2 days) of new services has become a key indicator for seizing new market
opportunities. However, when new services involve network expansion, the TTM lasts at least 1 to 2
months because purchase of boards, logistics, and installation are required. Once service provisioning
becomes time-consuming, a large number of market opportunities will be lost.
By enabling real-time resource visualization and prediction and using AI based algorithm to perform
rolling forecast, the resource assurance package of Huawei Optical Intelligence supports resource
purchase within 1 to 3 months and rolling budget within 6 to 12 months in advance. In addition, it
automatically discovers resource bottlenecks through online planning, provides accurate guidance for
network capacity expansion, and enables resources to be ready in advance, thereby shortening the TTM
to hours.
According to analysis and evaluation jointly conducted by Huawei and a carrier in 2019, the resource
assurance function package reduced the resource check time from 3 weeks to 1 hour, improving the
check efficiency by 98%. The TTM was reduced from 2 months to 2 hours, and the annual revenue
could increase by over USD 0.5 million.
4.2 Resource Assurance, Automatic Resource Bottleneck Discovery, and Precise Capacity Expansion
Planning data RMS
NCE(Live-network
data)
Centralized resource visualization:
network topology, sites and devices,
slots and ports, link and wavelength
usage, service trend, and service
distribution.
Resource comparison, conflict
analysis, and resource combination
based on planning data and
live-network data. In addition,
resource deployment status and
planning status can be
differentiated.
Historical
Forecast
Predictedservicematrix
1. Service matrix of
the upper-layer
network
3. Intelligent prediction based on historical growth
2. Service prediction based on historical growth
rate
Capacity threshold
$=?
New links and boards
Based on the rolling forecast of
historical data of over 12 months,
board procurement of 1 to 3
months and the rolling budget of 6
to 12 months can be supported.
Online capacity expansion planning
is supported and results are
delivered to the NMS, improving
configuration efficiency. Data
includes OCh and fiber parameters.
Planning Design Simulation
Service prediction
PO placement
Budget plan
New service
Routine planning
(OChcapacity
expansion)
Deployment
Deployment
Deployment
Resource visualization Capacity prediction Online planning
Zero-Touch MaintenanceHuawei Optical IntelligenceWhite Paper
Figure 4-4 High-reliability GMPLS/ASON networks
With the popularization of mesh networking, traditional networks face a number of challenges in
service reliability (fiber faults) and O&M efficiency (dependency on manual operations). The
GMPLS/ASON technology loads the intelligent brain (control plane) on the network to implement
automatic resource discovery and service rerouting. In this way, networks are evolving towards
automation and intelligence. Network meshing and ASON rerouting restoration change one-time
service protection to service connectivity upon path availability, achieving reliability of 99.999%.
Moreover, the automatic recovery capability of rerouting in seconds will greatly improve customer
experience and O&M efficiency.
4.3 Highly Reliable Network, Improving Service Reliability
In early 2019, Huawei and an European Tier-1 carrier successfully completed joint innovation test on
optical network health assurance. After collecting 2-month data from the live network and using the
AI-powered optical network health prediction tool to analyze and predict data, it was found that 200+
OTSs and OChs were in sub-healthy or faulty state, 52% of which were in sub-healthy state. Warnings
were generated more than 2.5 days in advance, and the check accuracy was higher than 90% when
compared with the actual result of the next month.
Customer's comments:
AI will be an important feature of automatic O&M in the future. By identifying and predicting risks,
proactive O&M can save 20% O&M labor costs for fiber faults.
4.4 Success Stories
Highreliability
EfficientO&M
ExtensiveSLAs
High cost-effectiveness
ASONcontrolplane
Qx
CCI NMI
Transport plane
Managementplane
99.999%Rerouting against multiple fiber cuts, reducing
service loss incurred from fiber cuts.
24/7 -> 8/5Automatic service rerouting and restoration,
easing O&M pressure.
5-level SLAsMulti-level SLAs, suiting various service
scenarios.
20%Resource sharing by rerouting and restoration,
cost-effective & reliable network.
4.4.1 Huawei and an European Tier-1 Carrier Successfully Completed the First AI-based Intelligent O&M and Joint Test for an Optical Network
14
Zero-Touch MaintenanceHuawei Optical Intelligence
White Paper
A carrier's network is located in an island. Due to frequent natural disasters such as earthquake and
tsunami, network reliability was poor (dozens of fiber cuts occurred every month). Traditional static 1+1
protection was used on the network, which had a high cost and low reliability. However, high-end
customers had high requirements on the network reliability. Service interruption would cause severe
penalties.
Network mesh reconstruction, electrical-layer ASON, and rerouting recovery capability against multiple
fiber cuts greatly improve service reliability from 98.5% to 99.99%, reduce the annual downtime from
130 hours to 1 hour, and lower the annual fault compensation by over USD 40 million.
15
Figure 4-5 Prediction result of a carrier
Figure 4-6 High-reliability network case
4.4.2 Highly Reliable Network Solution with Reliability Improved from 98.5% to 99.99%, Saving Annual Cost by 40M+
Shortening service downtime= Saving compensation costs
As Is: Without ASON
Platinum Gold Ratio of Penaltyto Rental
<28min <40min 5%
1h 4h 10%
3h 12h 30%
5h 20h 50%
6h 24h 100%
Static 1+1 protection
To Be: With ASON 2.0
Dynamic protection againstmultiple fiber cuts
Reliability: 98.5%Annual failure duration: 130 hoursCompensation amount: 44.98M
Reliability: 99.99%Annual failure duration: 0.8 hoursCompensation amount: 0.3M(revenue increase > 40M)
Prospects05
ProspectsHuawei Optical IntelligenceWhite Paper
Abraham Maslow, an American psychologist, classified human needs into five levels from low to high:
Physiological needs
Safety needs
ProspectsHuawei Optical Intelligence
White Paper
Figure 5-1 Five levels of customer requirements for transport software solutions
Love and belonging
Esteem
Self-actualization
Similarly, in Huawei Optical Intelligence, customers’ requirements for transport software solutions can
also be divided into five levels, as shown in the following figure.
In the future, Huawei Optical Intelligence aims to build E2E lifecycle automation, realize network
autonomy, enable IT and CT convergence, provide the ultimate user experience, raise revenue and
reduce expenditures, maximize business value, and finally help customers achieve business success.
· Unattended NM centers· Nearly zero resource waste· Always-online services
· Predictable resources, deployment upon planning· Predictable network heath and rerouting
· Open NBIs on NMS, cross-vendor management· Flexible optical-layer grooming, remote service provisioning
· P2P WDM· Manual network
· Service quality: fast service provisioning, latency map, guaranteed bandwidth on demand, customer self-management· Open and programmable, customized portal, improving tenant experience
IoT
5G + Cloud
4G + Private line
3G + Video
2G + Voice Static connection
Connection automation
Service automation
Intelligence
Networkautonomy
Driving Force Optical Network
Automation HierarchyTypical Demand
16
Huawei Technologies Co., Ltd.Huawei Industrial Base, Bantian, LonggangShenzhen, ChinaTel:+86 755 28780808Zip code:518129www.huawei.com