SlideShare a Scribd company logo
2
Most read
4
Most read
5
Most read
U2020
Virtualization
Increasing
equipment usage
and reducing costs
Simplifying
engineering
deployment and
capacity expansion
Extending the
flexibility of service
deployment
Facilitating load
balancing
VMs
Physical hosts
Hardware: physical hosts, storage devices, and
switching devices
Virtualization software
Infrastructure layer
Virtualization layer
Virtualization software defines and divides IT resources, including
CPUs, memories, disks, network adapters, and application programs. It
dynamically allocates and schedules IT resources for VMs, increasing
the resource usage.
Server virtualization refers to the creation of multiple logical computers
(virtual machines or VMs) on a single computer (physical host). Multiple
VMs can run simultaneously on a physical host and run different
operating systems. Application programs on each VM run independently.
Server virtualization improves the working efficiency of the physical host.
App
OS
App
OS+DB
App
OS+DB
App
OS
Virtual resources: CPUs, memories, network
adapters, and disks
Product Version: V300R019C10
Issue: 01
Virtualization Solution Overview (EulerOS, TaiShan)
Virtualization Overview1
Virtualization software centrally manages storage
volumes, snapshots, and volume mirrors.
EMC IBM HP HW
Storage pool and LUN sharing
SAN
Storage virtualization refers to the addition of a virtualization layer between the
physical storage systems and the servers. All storage systems are managed at this layer.
Storage virtualization isolates the servers from storage hardware. Adding, deleting,
replacing, splitting, or combining storage hardware is imperceptible to the server side.
Virtual disk
Virtual disk
Virtual disk Virtual disk
Virtual disk
Network virtualization refers to the integration of various physical networks into a
virtual network based on server virtualization. Services are able to run on the virtual
network. The physical networks integrated include Ethernet, FC SAN, and IP SAN. The
virtual network detects and automatically adapts to VM changes, for example, migrating
or copying a VM.
Physical storage
Virtual storage
Physical host
VMs
Virtualservice
network1
Virtualservice
network2
Virtualmaintenance
network1
Virtualmaintenance
network2
Virtualdatanetwork
1
Virtualdatanetwork
2
Physical NICs
Network switching center
(Ethernet, FC SAN, and IP
SAN)
Virtual Network
Active Plane
Virtual Network
Standby Plane
Virtualservice
network1
Virtualservice
network2
Virtualmaintenance
network1
Virtualmaintenance
network2
Virtualdatanetwork
1
Virtualdatanetwork
2
Virtual NICs Virtual NICs
Network
virtualization
Physical networks
Customer's
network
SANVirtual Disk
Virtual Disk
Virtual Disk
Virtual Disk
Virtual Disk
Virtual storage
EMC IBM HP HW
Storage Pool LUN Sharing
Physical storage
Virtualization software centrally manages storage
volumes, snapshots, and volume mirrors.
If the storage device
changes, applications
on the physical host
are not interrupted.
Hot Migration of VMs
Enables:
System capacity expansion
without VM power-off
Load balancing between physical
hosts in a distributed system
Hardware maintenance without
VM power-off
Lossless upgrades
X
Encapsulation: Each VM is saved in
files. Users can move or copy VMs by
moving or copying the corresponding
files.
Independence: VMs are independent
from hardware and can run on servers
with required resources, without any
modification.
Subarea: Different subareas are
allocated to VMs so that these VMs can
run simultaneously on a physical host.
Isolation: The VMs running on a
physical host are isolated from
each other.
Physical hosts can function as standby hosts for one another. A physical host does not need
to be exclusively configured as a standby host.
Compared with traditional deployment solutions, the virtualization solution allows for an
increased usage of existing resources, as well as enabling the physical host to handle an
increased load.
When faulty hosts are recovered, the system migrates the VMs back in seconds using the
hot migration technology. This does not interrupt services because the client and NEs do not
detect the migration.
Compared with traditional HA technologies, the operating system (OS) needs to be restarted
in the HA. The HA increases the time needed for VM startup and switchover. The increased
time does not exceed 5 minutes.
Service failures on VMs cannot be rectified by using HA.
High Availability (HA)
If the system detects faults in one or
multiple physical hosts, it migrates
the VMs from faulty physical hosts to
normal physical hosts and restarts
the VMs.
Hot Migration of Storage
Eliminates the need of VM power-off
Supports optical fiber channels
Key Virtualization Technologies2
App
X
App
OS
App
OS
App
OS
App
OS
App
OS
App
OS
Normal running during hot migration
U2020 Virtualization Solution3
 FusionSphere: is a Huawei cloud OS.
 OSMU: is short for Virtual OSS Self-maintenance Unit. It centrally manages IP addresses,
routes, time, time zones, data backup, the life cycle of OSS VMs, and data restoration.
 U2020: is an NE management system.
 ES: refers to the virtual emergency system. It can be interconnected to only one U2020
virtual system. For details on the service solution, see U2020 Virtual Emergency System
User Guide (EulerOS, TaiShan).
 TS: refers to the virtual Trace Server system. It can be interconnected to only one U2020
virtual system. For details on the service solution, see U2020 Trace Server Virtual System
User Guide (EulerOS, TaiShan).
Architecture and Application Scenarios
Notes
Centralized VM O&M
The OSMU facilitates centralized VM operation and maintenance (O&M). Use the OSMU to
operate and maintain those VMs connected to the OSMU, as shown in the following figure.
Data center
including servers,
storage devices,
network devices,
and virtualization
software.
Hardware: physical hosts, storage devices, and switching
devices
Virtualization software: FusionSphere
Virtual resources: CPUs, memories, network adapters,
and disks
OSMU
OS
U2020
OS+DB
Pre-integration: is an end-to-end OSS virtualization solution in which the OSS or NE solutions
provides hardware, virtualization software, VMs, and service software.
ES
OS+DB
TS
OS
Network
Level-1
Scenario
Level-2 Scenario
Virtualization
Software
Hardware
VM Deployment
Mode
OSS Product
Wireless
network
Pre-
integration
Integration by
OSS's
FusionSphere Data
Center
FusionSphere
OpenStack
TaiShan Pre-installation
OSMU/U2020/
ES/TS/mAOS
Core
network
Pre-
installation
CloudCore pre-
installation
FusionSphere
OpenStack
TaiShan
Use the Swift
Deploy tool to
create the OSMU.
And then use the
OSMU to create
the other OSS
products VMs.
OSMU/U2020
CloudEdge pre-
installation
FusionSphere
OpenStack
TaiShan OSMU/U2020
Full
installation
CloudCore full
installation
FusionSphere
OpenStack
TaiShan
OSMU/U2020/
ES
CloudEdge full
installation
FusionSphere
OpenStack
TaiShan
OSMU/U2020/
ES
Data Type Data Description Backup Policy
OS data Data of the VM operating
system
A maximum of 3 backup files containing operating system data
can be saved by default.
Static data Binary codes of the
OSMU software, OSS
applications, database
software, and third-party
software
•A maximum of 5 backup files containing database applications
can be saved by default.
•A maximum of 5 backup files containing OSS applications can
be saved by default.
•A maximum of 3 backup files containing the OSMU management
system can be saved by default.
Dynamic data Dynamic configuration
files and data generated
when the OSS is running
A maximum of 7 backup files containing dynamic data can be
saved by default.
When the number of backup files stored in the system exceeds
the configured threshold, the oldest backup files will be deleted
automatically.
1 Network HA
HA Overview
2 VM HA
3 Data HA
Switches, NICs, virtual switches, and vNICs can be configured in 1+1 redundancy mode to achieve network full
redundancy, improving network reliability.
RAID and multi-path technologies can be used to improve storage reliability.
Backup and Restoration on the OSMU
OSMU
OS
U2020
OS+DB
ES
OS+DB
TS
OS
Network switching center (Ethernet, FC SAN, and IP SAN)
OS data
Static data
Dynamic data
OSMU
TS
ES
U2020
OSMU
TS
ES
U2020
OSMU
TS
ES
U2020
Backup of dynamic data is not required for the ES. The
ES temporarily takes over U2020 services. When the
U2020 is recovered, dynamic data can be obtained in
the automatic supplementary collection procedure.
Use backup
files to restore
the system to
any status it
was in when it
was backed up.
TS U2020
U2020
U2020
OSMU
OSMU
OSMU
TS U2020
U2020
U2020
OSMU
OSMU
OSMU
TS U2020
U2020
U2020
OSMU
OSMU
OSMU
Pre-integration:
backup in NFS
mode
By-integration:
backup in SFTP
mode
Data transmission
layer
Backup and restoration
instruction layer
TS
TS
TS
TS
TS
TS
Backup Storage
A. HA of key service unit deployment policies
Multiple FusionSphere control nodes are deployed to balance load. In normal cases, all nodes provide services
simultaneously. If a node is faulty, the services on this node are automatically shared by other nodes that are
running properly, and therefore services are not affected.
Two FusionSphere OpenStack OM VMs are deployed on different control nodes.
One standby service VM and one standby database VM are deployed in a U2020 system. If a service or database
VM becomes faulty, services on the faulty service or database VM can be automatically switched to the standby
service or database VM, respectively. After the switchover, the standby VM uses the IP address of the faulty VM.
The emergency system and primary system are deployed on different hardware in anti-affinity mode. The two
systems are completely isolated from each other to prevent their concurrent faults caused by a single-point
hardware failure. The emergency system data is stored on a local hard disk, and the primary system data is stored
on disk arrays.
The emergency system, standby for the primary one, does not implement HA. If emergency system hardware is
faulty, the emergency system cannot be migrated to other devices.
NOTE: The Core Network does not support the standby VM solution.
B. For details about HA and hot migration technologies, see Key Virtualization Technologies.
Copyright © Huawei Technologies Co., Ltd.

More Related Content

PDF
Actix analyzer lte training
PDF
CS-Core Mobile Network (General)
PPT
Bts installation & commisioning.ppt
PPTX
5g-Air-Interface-pptx.pptx
PPTX
Lte default and dedicated bearer / VoLTE
PPTX
2G / 3G / 4G / IMS / 5G Overview with Focus on Core Network
PPTX
Introduction to Nokia RNC
PDF
Intermediate: 5G Network Architecture Options (Updated)
Actix analyzer lte training
CS-Core Mobile Network (General)
Bts installation & commisioning.ppt
5g-Air-Interface-pptx.pptx
Lte default and dedicated bearer / VoLTE
2G / 3G / 4G / IMS / 5G Overview with Focus on Core Network
Introduction to Nokia RNC
Intermediate: 5G Network Architecture Options (Updated)

What's hot (20)

PPTX
5gc call flow
PDF
Lte principle
PPTX
Technical_Training_of_5G_Networking_Design.pptx
PDF
Digital Mobile Network Evolution - from GSM to 5G
PDF
Ncat ccna cheat sheet
PDF
5G network architecture progress
PPT
Ip ran v1.1
PDF
1_WTTx LTM Main Slide_Huawei_Presentation.pdf
PDF
Interworking wcdma to lte
PDF
Advanced: Control and User Plane Separation of EPC nodes (CUPS)
PDF
5 g core network and the cloud - A standards perspective
PDF
Ttalteoverview 100923032416 Phpapp01 (1)
PPTX
3GPP Packet Core Towards 5G Communication Systems
PDF
6G Training Course Part 9: Course Summary & Conclusion
PDF
Nokia FLexi Edge BTS Installation Cabling and Checks
PDF
5G technical_overview_training_sec_1
PDF
Huawei White Spaces E & V Band Technology
PPT
2 g training optimization
PDF
Huawei ipran solution
PPTX
Sharing session huawei network optimization january 2015 ver3
5gc call flow
Lte principle
Technical_Training_of_5G_Networking_Design.pptx
Digital Mobile Network Evolution - from GSM to 5G
Ncat ccna cheat sheet
5G network architecture progress
Ip ran v1.1
1_WTTx LTM Main Slide_Huawei_Presentation.pdf
Interworking wcdma to lte
Advanced: Control and User Plane Separation of EPC nodes (CUPS)
5 g core network and the cloud - A standards perspective
Ttalteoverview 100923032416 Phpapp01 (1)
3GPP Packet Core Towards 5G Communication Systems
6G Training Course Part 9: Course Summary & Conclusion
Nokia FLexi Edge BTS Installation Cabling and Checks
5G technical_overview_training_sec_1
Huawei White Spaces E & V Band Technology
2 g training optimization
Huawei ipran solution
Sharing session huawei network optimization january 2015 ver3
Ad

Similar to U2020 virtualization solution overview (eular os, taishan) (20)

PDF
Transcending Computing Environment Boundaries: Seamless Computing Environmen...
PDF
Virtualization in Cloud Computing
PPTX
Virtualization Cloud computing technology
PPTX
VMware Overview
PDF
CPU Performance in Data Migrating from Virtual Machine to Physical Machine in...
PPT
Getting Started With Virtualization
PPTX
What is coming for VMware vSphere?
PPTX
cloud_and_virtualization_concepts by Dr Sudha L K Associate Professor BIT Ban...
PPTX
Microsoft Server Virtualization and Private Cloud
PPTX
Server Virtualization using Hyper-V
DOCX
VIRTUALIZATION FOR DATA CNTER AUTOMATION.docx
PPT
An Introduction To Server Virtualisation
PDF
Introduction to Virtualization
PPTX
SPPU_TE_COMPUTER_CLOUD_COMPUTING_unit 3.pptx
PPTX
sppu_TE_Comp_Cloud_computing_unit 3_cc.pptx
PPTX
Cloud computing virtualization
PPTX
cloud basics.
PPTX
Server Virtualization Sentrana Seminar
PPT
Fulcrum Group Virtualization How does It Fit
PPTX
Cloud computing virtualization
Transcending Computing Environment Boundaries: Seamless Computing Environmen...
Virtualization in Cloud Computing
Virtualization Cloud computing technology
VMware Overview
CPU Performance in Data Migrating from Virtual Machine to Physical Machine in...
Getting Started With Virtualization
What is coming for VMware vSphere?
cloud_and_virtualization_concepts by Dr Sudha L K Associate Professor BIT Ban...
Microsoft Server Virtualization and Private Cloud
Server Virtualization using Hyper-V
VIRTUALIZATION FOR DATA CNTER AUTOMATION.docx
An Introduction To Server Virtualisation
Introduction to Virtualization
SPPU_TE_COMPUTER_CLOUD_COMPUTING_unit 3.pptx
sppu_TE_Comp_Cloud_computing_unit 3_cc.pptx
Cloud computing virtualization
cloud basics.
Server Virtualization Sentrana Seminar
Fulcrum Group Virtualization How does It Fit
Cloud computing virtualization
Ad

More from kpphelu (10)

DOCX
MS Word Tip: Character to symbol conversion in Microsoft Word
PDF
NTC Question level 8 open 2074 chaitra
PDF
Ch 1 optical fiber introduction
PDF
Nt financial regulation 2072index
PDF
Syllabus for Level 7 to Level 8
PDF
Tu 39th convocation
PDF
Radio installation
PDF
Radio installation
DOCX
KPI measurement
PDF
Nijamati sewa-puraskar-paripatra
MS Word Tip: Character to symbol conversion in Microsoft Word
NTC Question level 8 open 2074 chaitra
Ch 1 optical fiber introduction
Nt financial regulation 2072index
Syllabus for Level 7 to Level 8
Tu 39th convocation
Radio installation
Radio installation
KPI measurement
Nijamati sewa-puraskar-paripatra

Recently uploaded (20)

PDF
ENT215_Completing-a-large-scale-migration-and-modernization-with-AWS.pdf
PPTX
cloud_computing_Infrastucture_as_cloud_p
PDF
gpt5_lecture_notes_comprehensive_20250812015547.pdf
PDF
project resource management chapter-09.pdf
PPTX
TLE Review Electricity (Electricity).pptx
PDF
7 ChatGPT Prompts to Help You Define Your Ideal Customer Profile.pdf
PDF
Agricultural_Statistics_at_a_Glance_2022_0.pdf
PDF
Univ-Connecticut-ChatGPT-Presentaion.pdf
PDF
August Patch Tuesday
PDF
Building Integrated photovoltaic BIPV_UPV.pdf
PDF
DASA ADMISSION 2024_FirstRound_FirstRank_LastRank.pdf
PDF
Enhancing emotion recognition model for a student engagement use case through...
PDF
WOOl fibre morphology and structure.pdf for textiles
PDF
Assigned Numbers - 2025 - Bluetooth® Document
PPTX
SOPHOS-XG Firewall Administrator PPT.pptx
PPTX
A Presentation on Touch Screen Technology
PPTX
Tartificialntelligence_presentation.pptx
PDF
NewMind AI Weekly Chronicles - August'25-Week II
PDF
1 - Historical Antecedents, Social Consideration.pdf
PDF
A novel scalable deep ensemble learning framework for big data classification...
ENT215_Completing-a-large-scale-migration-and-modernization-with-AWS.pdf
cloud_computing_Infrastucture_as_cloud_p
gpt5_lecture_notes_comprehensive_20250812015547.pdf
project resource management chapter-09.pdf
TLE Review Electricity (Electricity).pptx
7 ChatGPT Prompts to Help You Define Your Ideal Customer Profile.pdf
Agricultural_Statistics_at_a_Glance_2022_0.pdf
Univ-Connecticut-ChatGPT-Presentaion.pdf
August Patch Tuesday
Building Integrated photovoltaic BIPV_UPV.pdf
DASA ADMISSION 2024_FirstRound_FirstRank_LastRank.pdf
Enhancing emotion recognition model for a student engagement use case through...
WOOl fibre morphology and structure.pdf for textiles
Assigned Numbers - 2025 - Bluetooth® Document
SOPHOS-XG Firewall Administrator PPT.pptx
A Presentation on Touch Screen Technology
Tartificialntelligence_presentation.pptx
NewMind AI Weekly Chronicles - August'25-Week II
1 - Historical Antecedents, Social Consideration.pdf
A novel scalable deep ensemble learning framework for big data classification...

U2020 virtualization solution overview (eular os, taishan)

  • 1. U2020 Virtualization Increasing equipment usage and reducing costs Simplifying engineering deployment and capacity expansion Extending the flexibility of service deployment Facilitating load balancing VMs Physical hosts Hardware: physical hosts, storage devices, and switching devices Virtualization software Infrastructure layer Virtualization layer Virtualization software defines and divides IT resources, including CPUs, memories, disks, network adapters, and application programs. It dynamically allocates and schedules IT resources for VMs, increasing the resource usage. Server virtualization refers to the creation of multiple logical computers (virtual machines or VMs) on a single computer (physical host). Multiple VMs can run simultaneously on a physical host and run different operating systems. Application programs on each VM run independently. Server virtualization improves the working efficiency of the physical host. App OS App OS+DB App OS+DB App OS Virtual resources: CPUs, memories, network adapters, and disks Product Version: V300R019C10 Issue: 01 Virtualization Solution Overview (EulerOS, TaiShan) Virtualization Overview1
  • 2. Virtualization software centrally manages storage volumes, snapshots, and volume mirrors. EMC IBM HP HW Storage pool and LUN sharing SAN Storage virtualization refers to the addition of a virtualization layer between the physical storage systems and the servers. All storage systems are managed at this layer. Storage virtualization isolates the servers from storage hardware. Adding, deleting, replacing, splitting, or combining storage hardware is imperceptible to the server side. Virtual disk Virtual disk Virtual disk Virtual disk Virtual disk Network virtualization refers to the integration of various physical networks into a virtual network based on server virtualization. Services are able to run on the virtual network. The physical networks integrated include Ethernet, FC SAN, and IP SAN. The virtual network detects and automatically adapts to VM changes, for example, migrating or copying a VM. Physical storage Virtual storage Physical host VMs Virtualservice network1 Virtualservice network2 Virtualmaintenance network1 Virtualmaintenance network2 Virtualdatanetwork 1 Virtualdatanetwork 2 Physical NICs Network switching center (Ethernet, FC SAN, and IP SAN) Virtual Network Active Plane Virtual Network Standby Plane Virtualservice network1 Virtualservice network2 Virtualmaintenance network1 Virtualmaintenance network2 Virtualdatanetwork 1 Virtualdatanetwork 2 Virtual NICs Virtual NICs Network virtualization Physical networks Customer's network SANVirtual Disk Virtual Disk Virtual Disk Virtual Disk Virtual Disk Virtual storage EMC IBM HP HW Storage Pool LUN Sharing Physical storage Virtualization software centrally manages storage volumes, snapshots, and volume mirrors. If the storage device changes, applications on the physical host are not interrupted.
  • 3. Hot Migration of VMs Enables: System capacity expansion without VM power-off Load balancing between physical hosts in a distributed system Hardware maintenance without VM power-off Lossless upgrades X Encapsulation: Each VM is saved in files. Users can move or copy VMs by moving or copying the corresponding files. Independence: VMs are independent from hardware and can run on servers with required resources, without any modification. Subarea: Different subareas are allocated to VMs so that these VMs can run simultaneously on a physical host. Isolation: The VMs running on a physical host are isolated from each other. Physical hosts can function as standby hosts for one another. A physical host does not need to be exclusively configured as a standby host. Compared with traditional deployment solutions, the virtualization solution allows for an increased usage of existing resources, as well as enabling the physical host to handle an increased load. When faulty hosts are recovered, the system migrates the VMs back in seconds using the hot migration technology. This does not interrupt services because the client and NEs do not detect the migration. Compared with traditional HA technologies, the operating system (OS) needs to be restarted in the HA. The HA increases the time needed for VM startup and switchover. The increased time does not exceed 5 minutes. Service failures on VMs cannot be rectified by using HA. High Availability (HA) If the system detects faults in one or multiple physical hosts, it migrates the VMs from faulty physical hosts to normal physical hosts and restarts the VMs. Hot Migration of Storage Eliminates the need of VM power-off Supports optical fiber channels Key Virtualization Technologies2 App X App OS App OS App OS App OS App OS App OS Normal running during hot migration
  • 4. U2020 Virtualization Solution3  FusionSphere: is a Huawei cloud OS.  OSMU: is short for Virtual OSS Self-maintenance Unit. It centrally manages IP addresses, routes, time, time zones, data backup, the life cycle of OSS VMs, and data restoration.  U2020: is an NE management system.  ES: refers to the virtual emergency system. It can be interconnected to only one U2020 virtual system. For details on the service solution, see U2020 Virtual Emergency System User Guide (EulerOS, TaiShan).  TS: refers to the virtual Trace Server system. It can be interconnected to only one U2020 virtual system. For details on the service solution, see U2020 Trace Server Virtual System User Guide (EulerOS, TaiShan). Architecture and Application Scenarios Notes Centralized VM O&M The OSMU facilitates centralized VM operation and maintenance (O&M). Use the OSMU to operate and maintain those VMs connected to the OSMU, as shown in the following figure. Data center including servers, storage devices, network devices, and virtualization software. Hardware: physical hosts, storage devices, and switching devices Virtualization software: FusionSphere Virtual resources: CPUs, memories, network adapters, and disks OSMU OS U2020 OS+DB Pre-integration: is an end-to-end OSS virtualization solution in which the OSS or NE solutions provides hardware, virtualization software, VMs, and service software. ES OS+DB TS OS Network Level-1 Scenario Level-2 Scenario Virtualization Software Hardware VM Deployment Mode OSS Product Wireless network Pre- integration Integration by OSS's FusionSphere Data Center FusionSphere OpenStack TaiShan Pre-installation OSMU/U2020/ ES/TS/mAOS Core network Pre- installation CloudCore pre- installation FusionSphere OpenStack TaiShan Use the Swift Deploy tool to create the OSMU. And then use the OSMU to create the other OSS products VMs. OSMU/U2020 CloudEdge pre- installation FusionSphere OpenStack TaiShan OSMU/U2020 Full installation CloudCore full installation FusionSphere OpenStack TaiShan OSMU/U2020/ ES CloudEdge full installation FusionSphere OpenStack TaiShan OSMU/U2020/ ES
  • 5. Data Type Data Description Backup Policy OS data Data of the VM operating system A maximum of 3 backup files containing operating system data can be saved by default. Static data Binary codes of the OSMU software, OSS applications, database software, and third-party software •A maximum of 5 backup files containing database applications can be saved by default. •A maximum of 5 backup files containing OSS applications can be saved by default. •A maximum of 3 backup files containing the OSMU management system can be saved by default. Dynamic data Dynamic configuration files and data generated when the OSS is running A maximum of 7 backup files containing dynamic data can be saved by default. When the number of backup files stored in the system exceeds the configured threshold, the oldest backup files will be deleted automatically. 1 Network HA HA Overview 2 VM HA 3 Data HA Switches, NICs, virtual switches, and vNICs can be configured in 1+1 redundancy mode to achieve network full redundancy, improving network reliability. RAID and multi-path technologies can be used to improve storage reliability. Backup and Restoration on the OSMU OSMU OS U2020 OS+DB ES OS+DB TS OS Network switching center (Ethernet, FC SAN, and IP SAN) OS data Static data Dynamic data OSMU TS ES U2020 OSMU TS ES U2020 OSMU TS ES U2020 Backup of dynamic data is not required for the ES. The ES temporarily takes over U2020 services. When the U2020 is recovered, dynamic data can be obtained in the automatic supplementary collection procedure. Use backup files to restore the system to any status it was in when it was backed up. TS U2020 U2020 U2020 OSMU OSMU OSMU TS U2020 U2020 U2020 OSMU OSMU OSMU TS U2020 U2020 U2020 OSMU OSMU OSMU Pre-integration: backup in NFS mode By-integration: backup in SFTP mode Data transmission layer Backup and restoration instruction layer TS TS TS TS TS TS Backup Storage A. HA of key service unit deployment policies Multiple FusionSphere control nodes are deployed to balance load. In normal cases, all nodes provide services simultaneously. If a node is faulty, the services on this node are automatically shared by other nodes that are running properly, and therefore services are not affected. Two FusionSphere OpenStack OM VMs are deployed on different control nodes. One standby service VM and one standby database VM are deployed in a U2020 system. If a service or database VM becomes faulty, services on the faulty service or database VM can be automatically switched to the standby service or database VM, respectively. After the switchover, the standby VM uses the IP address of the faulty VM. The emergency system and primary system are deployed on different hardware in anti-affinity mode. The two systems are completely isolated from each other to prevent their concurrent faults caused by a single-point hardware failure. The emergency system data is stored on a local hard disk, and the primary system data is stored on disk arrays. The emergency system, standby for the primary one, does not implement HA. If emergency system hardware is faulty, the emergency system cannot be migrated to other devices. NOTE: The Core Network does not support the standby VM solution. B. For details about HA and hot migration technologies, see Key Virtualization Technologies. Copyright © Huawei Technologies Co., Ltd.