- Released At: 22-10-2024
- Page Views:
- Downloads:
- Table of Contents
- Related Documents
-
|
AD-DC 6.5 |
Unified Platform and Components Deployment Guide |
|
|
Document version: 5W101-20241021
Copyright © 2024 New H3C Technologies Co., Ltd. All rights reserved.
No part of this manual may be reproduced or transmitted in any form or by any means without prior written consent of New H3C Technologies Co., Ltd.
Except for the trademarks of New H3C Technologies Co., Ltd., any trademarks that may be mentioned in this document are the property of their respective owners.
This document provides generic technical information, some of which might not be applicable to your products.
The information in this document is subject to change without notice.
Hardware resource requirements
About cluster separated deployment
About cluster converged deployment
Deploy SeerEngine-DC, vBGP, DTN, and Super Controller
Deploy SeerAnalyzer and collectors
Deploy the NSM network management component
Appendix A Obtain documentation
Overview
Components
This document describes the deployment procedures for the following AD-DC components:
· Unified Platform—Platform component on which you deploy controllers and analyzers.
· SeerEngine-DC—Controller component that runs SDN applications. It controls resources on the network and is the network management center.
· SeerAnalyzer—Analyzer component that provides intelligent analysis for AD-DC solutions.
· vBGP—Virtual BGP component that provides conversion between host overlay flow tables and network overlay EVPN routes. It is typically used in hybrid overlay scenarios.
· Digital-Twin-Network (DTN)—Simulation component that provides the simulation functionality.
· Collector—Collects and sends traffic to SeerAnalyzer.
· Super Controller—Manages the SDN controller cluster in the multi-data center scenario and provides unified service orchestration and O&M across data centers.
Deployment workflow
Figure 1 Deployment workflow
Hardware resource requirements
For the server hardware resources required to install the components, see AD-NET Solution Hardware Configuration Guide.
Plan network configuration
The following information uses examples to describe how to plan network configuration for Unified Platform and all its components. You can select components for your deployment scenario as required.
Super Controller uses the northbound service VIP of Unified Platform to communicate with sites. You do not need to plan a separate management IP address for it.
The vBGP component supports both the management network and service network converged scheme and management network and service network separated scheme. This document uses the management network and service network converged scheme for the controller for example and the southbound and northbound integrated network (no southbound network) scheme for the analyzer for example.
This section plans only for the management network configuration. For service network configuration, see the network configuration plan for the specific deployment scenario.
The IP address plan in this section is for illustration only.
Standalone deployment
About standalone deployment
The vBGP component is not supported in the standalone deployment scenario.
In a standalone deployment, you deploy SeerEngine-DC, Super Controller, and SeerAnalyzer on separate Unified Platform nodes, and deploy the collector on a server other than the Unified Platform nodes.
Connect all the Unified Platform nodes to the management switch and configure the management switch to provide gateway services.
IP address plan
Component |
IP address type |
IP addresses |
Remarks |
Unified Platform 1 (on which SeerEngine-DC is deployed) |
Master node IP |
192.168.10.102/24 |
These IP addresses must be on the same network segment. Specify gateway 192.168.10.1, which is placed on the management switch. |
Cluster internal VIP |
192.168.10.101/32 |
||
Northbound service VIP |
192.168.10.100/32 |
||
Unified Platform 2 (on which SeerAnalyzer is deployed) |
Master node IP |
192.168.10.202/24 |
These IP addresses must be on the same network segment. Specify gateway 192.168.10.1, which is placed on the management switch. |
Cluster internal VIP |
192.168.10.201/32 |
||
Northbound service VIP |
192.168.10.200/32 |
||
Unified Platform 3 (on which Super Controller is deployed) |
Master node IP |
192.168.10.212/24 |
Super Controller uses the northbound service VIP of Unified Platform to communicate with sites. You do not need to plan a separate management IP address for it. |
Cluster internal VIP |
192.168.10.211/32 |
||
Northbound service VIP |
192.168.10.210/32 |
||
Collector |
Management IP |
192.168.10.50/24 |
Specify gateway 192.168.10.1, which is placed on the management switch. |
Southbound collection IP |
11.1.1.0/24 |
You need to configure two IP addresses. Use one to receive mirrored network device packets. Use the other as the SeerCollector float IP for it to be discovered by devices. |
|
SeerEngine-DC |
Management network |
Subnet: 192.168.12.0/24 |
Configure a MACVLAN-type management network. Specify gateway 192.168.12.1, which is placed on the management switch. |
SeerAnalyzer (collector) |
Management network |
Subnet: 192.168.12.0/24 |
Configure a MACVLAN-type management network. Specify gateway 192.168.12.1, which is placed on the management switch. |
Cluster separated deployment
About cluster separated deployment
In a cluster separated deployment, you deploy SeerEngine-DC, SeerAnalyzer (analyzer), and Super Controller in separate Uniform Platform clusters, and deploy the collector on a separate physical server.
This document uses the following cluster separated deployment for example:
· Unified Platform cluster 1—Contains three master nodes and one worker node for SeerEngine-DC deployment. vBGP is deployed on two master nodes, and DTN is deployed on the worker node.
· Unified Platform cluster 2—Contains three master nodes for SeerAnalyzer deployment.
· Unified Platform cluster 3—Contains three master nodes for Super Controller deployment.
· Separate server—Used for collector deployment.
Connect all these servers to the management switch and configure the management switch to provide gateway services. You must use a network interface on each of the two servers on which vBGP is deployed for configuring a management network for vBGP.
The Unified Platform cluster for SeerEngine-DC can span one subnet or multiple subnets. For high availability, you can deploy the master nodes in 2+1+1 mode. For information about address planning and available deployment modes, see H3C SeerEngine-DC Installation Guide. In this deployment example, the nodes are deployed on the same subnet.
IP address plan
Component |
IP address type |
IP addresses |
Remarks |
Unified Platform cluster 1 (on which SeerEngine-DC is deployed) |
Master node 1 IP |
192.168.10.102/24 |
In this example, all the IP addresses are on the same subnet. Specify gateway 192.168.10.1, which is placed on the management switch. |
Master node 2 IP |
192.168.10.103/24 |
||
Master node 3 IP |
192.168.10.104/24 |
||
Worker node 1 IP (DTN) |
192.168.10.105/24 |
||
Cluster internal VIP |
192.168.10.101/32 |
||
Northbound service VIP |
192.168.10.100/32 |
||
Unified Platform cluster 2 (on which SeerAnalyzer is deployed) |
Master node 4 IP (SeerAnalyzer) |
192.168.10.202/24 |
These IP addresses must be on the same network segment. Specify gateway 192.168.10.1, which is placed on the management switch. |
Master node 5 IP (SeerAnalyzer) |
192.168.10.203/24 |
||
Master node 6 IP (SeerAnalyzer) |
192.168.10.204/24 |
||
Cluster internal VIP |
192.168.10.201/32 |
||
Northbound service VIP |
192.168.10.200/32 |
||
Unified Platform cluster 3 (on which Super Controller is deployed) |
Master node 7 IP (Super Controller) |
192.168.10.212/24 |
These IP addresses must be on the same network segment. Specify gateway 192.168.10.1, which is placed on the management switch. |
Master node 8 IP (Super Controller) |
192.168.10.213/24 |
||
Master node 9 IP (Super Controller) |
192.168.10.214/24 |
||
Cluster internal VIP |
192.168.10.211/32 |
||
Northbound service virtual IP |
192.168.10.210/32 |
||
Collector |
Management IP |
192.168.10.50/24 |
Specify gateway 192.168.10.1, which is placed on the management switch. |
Southbound collection IP |
11.1.1.0/24 |
You need to configure two IP addresses. Use one to receive mirrored network device packets. Use the other as the SeerCollector float IP for it to be discovered by devices. |
|
SeerEngine-DC |
Management network |
Subnet: 192.168.12.0/24 |
Configure a MACVLAN-type management network. Specify gateway 192.168.12.1, which is placed on the management switch. |
vBGP |
Management network and service network converged |
Subnet: 192.168.13.0/24 |
Configure MACVLAN networks. Specify gateway 192.168.13.1, which is placed on the management switch. |
DTN |
Management network |
Subnet: 192.168.15.0/24 |
Configure a MACVLAN-type management network. Specify gateway 192.168.12.1, which is placed on the management switch. |
DTN host node |
Simulation management network |
Network address pool: 192.168.12.134/24 to 192.168.12.140/24 |
If the controller management network and simulation management network use the same management switch, you must also configure VPN instances for isolation on the management switch to prevent IP address conflicts from affecting the services. If the controller management network and simulation management network use different management switches, physically isolate these switches. |
Simulated device service network |
Network address pool: 192.168.11.134/24 to 192.168.11.140/24 |
Simulated device service network, which is on the same network segment as the production device service network |
|
Node management network |
Network address pool: 192.168.10.110 to 192.168.10.120 |
The default gateway is 192.168.10.1 on the management switch |
|
SeerAnalyzer (collector) |
Management network |
Subnet: 192.168.12.0/24 Network address pool: 192.168.12.141 to 192.168.12.172 |
Configure a MACVLAN-type management network. Specify gateway 192.168.10.1, which is placed on the management switch. |
Cluster converged deployment
About cluster converged deployment
In a cluster converged deployment, you deploy SeerEngine-DC, SeerAnalyzer, and Super Controller on one Unified Platform cluster.
This document uses the following cluster converged deployment for example:
· Deploy a Unified Platform cluster that has three master nodes and four worker nodes.
· Deploy SeerEngine-DC and Super Controller on the three master nodes.
· Deploy SeerAnalyzer and collector on two of the worker nodes.
· Deploy vBGP on two of the master nodes.
· Deploy DTN on one of the worker nodes.
· Deploy collectors on servers outside of the cluster. In this example, one collector is deployed.
Connect all these servers to the management switch and configure the management switch to provide gateway services. You must use a network interface on each of the two servers on which vBGP is deployed for configuring a management network for vBGP.
IP address plan
Component |
IP address type |
IP addresses |
Remarks |
Unified Platform |
Master node 1 IP |
192.168.10.102/24 |
These IP addresses must be on the same network segment. Specify gateway 192.168.10.1 for the component, which is placed on the management switch. Super Controller uses the northbound service VIP of Unified Platform to communicate with sites. You do not need to plan a separate management IP address for it. |
Master node 2 IP |
192.168.10.103/24 |
||
Master node 3 IP |
192.168.10.104/24 |
||
Worker node 7 IP (DTN) |
192.168.10.105/24 |
||
Worker node 4 IP (SeerAnalyzer) |
192.168.10.202/24 |
These IP addresses must be on the same network segment. Specify gateway 192.168.10.1, which is placed on the management switch. |
|
Worker node 5 IP (SeerAnalyzer) |
192.168.10.203/24 |
||
Worker node 6 IP (SeerAnalyzer) |
192.168.10.204/24 |
||
Cluster internal VIP |
192.168.10.101/32 |
||
Northbound service VIP |
192.168.10.100/32 |
||
Collector |
Management IP |
192.168.10.50/24 |
Specify gateway 192.168.10.1, which is placed on the management switch. |
Southbound collection IP |
11.1.1.0/24 |
You need to configure two IP addresses. Use one to receive mirrored network device packets. Use the other as the SeerCollector float IP for it to be discovered by devices. |
|
SeerEngine-DC |
Management network |
Subnet: 192.168.12.0/24 |
Configure a MACVLAN-type management network. Specify gateway 192.168.12.1 for the component, which is placed on the management switch. |
vBGP |
Management network |
Subnet: 192.168.13.0/24 |
Configure an MACVLAN-type management network. Specify gateway 192.168.13.1, which is placed on the management switch. |
DTN |
Management network |
Subnet: 192.168.15.0/24 |
Configure a MACVLAN-type management network. Specify gateway 192.168.15.1, which is placed on the management switch. |
DTN host node |
Simulation management network |
Network address pool: 192.168.12.134/24 to 192.168.12.140/24 |
If the controller management network and simulation management network use the same management switch, you must also configure VPN instances for isolation on the management switch to prevent IP address conflicts from affecting the services. If the controller management network and simulation management network use different management switches, physically isolate these switches. |
Simulated device service network |
Network address pool: 192.168.11.134/24 to 192.168.11.140/24 |
Simulated device service network, which is on the same network segment as the production device service network |
|
Node management network |
Network address pool: 192.168.10.110 to 192.168.10.120 |
The default gateway is 192.168.10.1 on the management switch |
|
SeerAnalyzer (collector) |
Management network |
Subnet: 192.168.12.0/24 |
Configure a MACVLAN-type management network. Specify gateway 192.168.10.1, which is placed on the management switch. The network port of a Worker node must be used exclusively for carrying the southbound network traffic. |
Deploy the components
Deploy Unified Platform
For the Unified Platform deployment procedure, see H3C Unified Platform Deployment Guide. For information about how to obtain this document, see "Appendix A Obtain documentation."
To run the components on Unified Platform, partition the system drive according to the requirements of each component, and deploy all necessary application installation packages. For more information about disk partitioning and required packages for each component, see the installation guide of each component.
Deploy SeerEngine-DC, vBGP, DTN, and Super Controller
For information about the SeerEngine-DC, vBGP, and DTN deployment procedures and required application installation packages, see H3C SeerEngine-DC Controller Installation Guide (Unified Platform). For information about how to obtain the document, see "Appendix A Obtain documentation".
For information about the Super Controller deployment procedure and required application installation packages, see H3C Super Controller Installation Guide (Unified Platform). For information about how to obtain the document, see "Appendix A Obtain documentation".
As a best practice, use the non-RDRS scheme for this solution.
Deploy SeerAnalyzer and collectors
For information about the SeerAnalyzer deployment procedure and required application installation packages, see H3C SeerAnalyzer Deployment Guide. For information about how to obtain the document, see "Appendix A Obtain documentation."
Follow these restrictions and guidelines when you deploy SeerAnalyzer and collectors:
1. To plan the network configuration, see network planning for single-stack southbound network in H3C SeerAnalyzer Deployment Guide.
2. After switching devices are deployed, use an interface on each collector to connect to a leaf device and change the NIC type to DPDK for traffic collection.
Deploy the NSM network management component
For more information about the installation and deployment procedure for the NSM network management component, see H3C NSM Installation Guide. For information about how to obtain the document, see "Appendix A Obtain documentation".
Table 7 shows the NSM network management component installation packages required by SeerEngine-DC.
Table 1 Description of application installation packages and dependencies
Installation package |
Installation package name |
Dependencies |
Network resource installation package |
NSM_FCAPS-Res_version.zip |
Required |
Network topology installation package |
NSM_FCAPS-Topo_version.zip |
Network resource installation package |
Network performance and monitoring installation package |
NSM_FCAPS-Perf_version.zip |
Network resource installation package |
Network management intelligent configuration center installation package |
NSM_FCAPS-ICC_version.zip |
Network resource installation package |
Network asset installation package |
NSM_FCAPS-Asset_version.zip |
Network resource installation package |
Device panel installation package |
NSM_FCAPS-Webdm_version.zip |
Network resource installation package |
Operations monitoring
For information about operations monitoring for this solution, see AD-DC 6.5 Operations Monitoring Configuration Guide.
Appendix A Obtain documentation
To obtain documentation for installing Unified Platform and its components:
1. Access the website at https://www.h3c.com/en/Support/Resource_Center/Technical_Documents/.
2. Select the product category and model and then obtain the document.
Use the following table to obtain the desired documents:
Product name |
Manual name |
Directory |
Unified Platform |
H3C Unified Platform Deployment Guide-E0715 |
AD-NET(SDN)/H3C SeerEngine-DC/Install & Upgrade |
SeerEngine-DC |
H3C SeerEngine-DC Installation Guide (Unified Platform )-E65xx |
AD-NET(SDN)/H3C SeerEngine-DC/Install & Upgrade |
SeerAnalyzer |
H3C SeerAnalyzer Deployment Guide-E65xx |
AD-NET(SDN)/H3C SeerAnalyzer-DC/Install & Upgrade |
Super Controller |
H3C Super Controller Installation Guide (Unified Platform)-E65xx |
AD-NET(SDN)/H3C Super Controller/Install & Upgrade |
NSM |
H3C NSM Installation Guide-E0715 |
Network Operations & Management/H3C U-Center 2.0 |