H3C ADDC 3.0 Director03-03-2020
Positioned as the next-generation data center management platform, the H3C Application-Driven Data Center (ADDC) Director provides service-oriented Operations and Management (O&M) capabilities for network resources from a unified GUI.
By integrating end-to-end operations of services, logical resources, and physical resources, the ADDC Director provides full lifecycle management for computing, storage, and network resources of the data center network and services. Based on the PDCA methodology, the ADDC Director provides O&M capabilities that cover network design, deployment, monitoring, and optimization.
Compared with traditional O&M platforms, the ADDC Director has the following characteristics:
Fully visualized data center planning and orchestration—Administrators can plan and orchestrate resources for setting up the data center network through the unified GUI. The ADDC Director provides visibility throughout the data center network planning and orchestration process. The ADDC Director ensures that the data center network planning and orchestration is inheritable, traceable, and deployable, decoupling IT capabilities from personnel capabilities.
Automated deployment—After completing network planning and orchestration, administrators can power on devices to trigger automated network deployment. The devices automatically retrieve their role-specific configuration files from the ADDC Director and configure themselves. The computing, storage, and network resources are automatically deployed on the network. This feature decouples configurations from devices and reduces the number of device configuration files that need to be maintained.
Fully visualized resource and service monitoring—The ADDC Director monitors the status and resource usage of managed devices and services from multiple perspectives and presents the monitoring data in various forms. Administrators can view the topologies of physical resources and logical resources, and the mappings among the physical resources, logical resources, and services.
Automatic fault detection and end-to-end troubleshooting—Through intelligent analysis of the multi-dimensional data collected from the data center network, the ADDC Director can automatically detect faults and perform end-to-end troubleshooting and analysis to pinpoint the fault causes.
Automatic application-driven optimization—The ADDC Director provides application-oriented closed-loop O&M capabilities for applications and services. It can actively and passively detect application and service anomalies and takes responsive actions. It also supports dynamic end-to-end resource scheduling to tackle application and service quality degradations.
Features and benefits
Visualized data center capacity planning and resource orchestration
The ADDC Director provides a unified GUI for administrators to centrally plan and orchestrate network and computing resources on the data center network. The global resource topology can be easily constructed with simple drag-and-drop operations to assist in global data center resource capacity and configuration planning.
The data center network planning process is divided into Plan Capacity, Initial Configuration, and Plan Topology phases.
In the Plan Capacity phase, the administrator first selects a networking scheme and then proceeds with the network capacity planning and calculation. Based on the spine-leaf architecture of the ADDC fabric, the ADDC Director provides three networking schemes (centralized VXLAN, distributed VXLAN, and VLAN) for setting up a data center network.
In the Initial Configuration phase, the administrator completes these configurations: DDI Configuration Management, Network Automation, and Server Automation.
Initial configuration process
DDI Configuration Management—Configure the DHCP server and IP address ranges for network management, server out-of-band management, and service ports.
Network Automation—Configure initial network device automation settings, including role-based device configuration files, automation parameters, automation policies, and device software versions. When devices are powered on, they obtain their respective configuration files based on their device role, position, model, and MAC address and configure themselves.
Server Automation—Configure initial automation settings for bare metal servers, including the configuration files to be delivered to the servers and automation parameters such as iLO and IPMI parameters. When bare metal servers are powered on, they obtain their respective configuration files based on their server model, access position, and serial number and configure themselves.
In the Plan Topology phase, the administrator configures the network topology by adding device nodes and links to the topology. After devices are powered on, the administrator can view the automated deployment progress and deployment result on the topology page.
Automated data center resource deployment
After finishing data center network planning and resource orchestration, automated network deployment can be started at any time as soon as the devices and servers are correctly installed and cabled.
Once powered on, the devices obtain the address of the TFTP server (embedded in the ADDC Director) from the DHCP server. Then, they download their role-specific configuration files from the TFTP server to configure themselves, install software as requested, and restart to complete automated network deployment and setup. The deployment process requires no user intervention and improves the deployment efficiency by nearly 100 times compared with the traditional manual deployment mode.
Automated data center resource deployment process
The devices are managed by the ADDC Director once they complete the automated deployment process. The ADDC Director displays managed devices on the Plan Topology page. From the page, administrators can implement maintenance tasks such as device and link auto discovery, VXLAN deployment, ACL management, configuration backup, software upgrade, and status monitoring on the devices.
Configuring devices after automated deployment
After automated network deployment is complete, administrators can proceed to configure fabric services such as underlay, overlay (VXLAN), and service chain services.
Visualized data center resource and service monitoring
The ADDC Director provides multi-dimensional monitoring capabilities for fabric resources (both physical and virtual) and services on the data center network.
Fabric resource monitoring—The ADDC Director monitors various types of fabric resources, including underlay and overlay networks, storage devices, servers, and virtual devices. Fabric resource monitoring can be further divided into the following types:
Underlay network monitoring—Allows administrators to view the following information:
Operating status of the controller and network devices, including their CPU, memory, and hard disk usage statistics.
Configuration and software versions on the network devices.
Faults detected on the devices.
Connectivity status of the devices. The ADDC Director provides a tool to test connectivity to managed devices.
Traffic statistics of the underlay network.
Global network topology.
Overlay network monitoring—Allows administrators to view the VXLAN configuration information and traffic information for VXLANs or tenant VNI tunnels.
Virtual device management and monitoring—The ADDC Director supports virtualization platforms of various vendors, including VMware, KVM, Hyper-V, Xen, and CAS. Administrators can create, update, configure, clone, and migrate VMs, create automation templates for VMs, and view VM monitoring information.
Service monitoring—To ensure the quality of services provided to users, the ADDC Director supports monitoring services from various perspectives, including the underlying applications on which services depend, traffic of the applications and services, and the overall service health level.
Application monitoring—The ADDC Director actively monitors the applications that services depend on such as the OS, database, and middleware on servers and VMs. The ADDC Director displays the monitoring information through combined network and application topologies, which helps administrators learn the applications' availability and health status in real time and enables timely anomaly detection.
Traffic monitoring—The ADDC Director monitors the traffic of applications and services and enables traffic analysis based on application type, bandwidth usage, duration, traffic source (from inside or outside the data center network), and traffic trend. The information helps administrators make informed decisions on bandwidth guarantee configuration and capacity planning for the data center network.
Service monitoring—The ADDC Director analyzes the overall health status of services by integrating the operational status of the underlying resources and applications used by the services from various perspectives. It evaluates the health level of the services based on a preset evaluation mechanism and presents the data on the GUI in a simple and easy-to-understand form. This greatly improves the O&M efficiency.
Automatic inspection and intelligent diagnosis
The ADDC Director provides multi-dimensional inspection and diagnosis tools that liberate the O&M administrators from the repetitive work of daily inspection. It supports both network inspection in virtual reality mode and scheduled device inspection based on user-defined inspection policies.
Network inspection in virtual reality mode—The O&M administrator can walk around and inspect the data center network through a simulated inspector in virtual reality mode. For example, the administrator can check the environment & power status, rack status, and device status in the equipment room without actually stepping into the equipment room.
Network inspection in virtual reality mode
Scheduled device inspection—The ADDC Director supports scheduled inspection on devices based on user-defined inspection policies. The O&M administrator defines inspection items, execution time, and anomaly judgment criteria in an inspection policy. The ADDC director periodically executes the policy to check hardware and software status, load condition, protocol status, and security status of devices, and generates a report for each inspection operation.
Automatic application-driven optimization
The ADDC Director provides application-oriented closed-loop O&M capabilities.
l Through means such as visualized monitoring, device and controller alarm analysis, and network inspection, the ADDC Director can automatically detect resource and service anomalies and take corresponding actions to resolve detected problems in a timely manner.
l Through intelligent analysis of performance, event, and traffic data, the ADDC Director can timely detect application and service quality degradation and implement end-to-end dynamic resource scheduling based on application-oriented response policies. The ADDC Director can also send notifications to administrators (through emails, SMS, voice alerts, and SNMP traps) so they can take corresponding actions.
Server requirements for ADDC Director
Available disk space
x86-64 (Intel 64/AMD 64), 24 cores or above, >=2.5 GHz
64 GB or above
512 GB or above Speed>=20MB/s
Windows Server 2008 (64bit)
Requires Patch Service Pack 2
Windows Server 2008 R2
Requires Patch Service Pack 1
Windows Server 2012
Requires Patch KB2836988
Windows Server 2012 R2
SQL Server 2008
Requires Patch Service Pack 3
SQL Server 2008 R2
Requires Patch Service Pack 2
SQL Server 2012
Requires Patch Service Pack 2
SQL Server 2014
Red Hat Enterprise Linux Server 5.5 (64bit)
Red Hat Enterprise Linux Server 5.9 (64bit)
Red Hat Enterprise Linux Server 6.1 (64bit)
Red Hat Enterprise Linux Server 6.4 (64bit)
Red Hat Enterprise Linux Server 7.x (64bit)
Oracle 11g Release 1
Oracle 11g Release 2
Oracle 12c Release 1
Virtual machine installation platform requirements
H3C recommends installing Fabric Director on a physical server. If Fabric Director is installed in a virtual machine, the deployment scenarios on the platform are as follows:
● VMware ESX + Windows
● VMware ESX + Linux
● Windows Hyper V + Windows
● H3C CAS + Windows
● H3C CAS + Linux
If the Fabric Director is installed on a virtual machine. Please do not modify the following virtual machine configuration parameters:
● Number of CPU cores
● Number of network cards, working mode and MAC address
● Number of disk drives, storage path
● Allocate storage space
Note: Modifying these parameters will cause Fabric Director does not work properly.
H3C DR1000 ADDC Fabric Director Software
H3C DR1000 ADDC Fabric Director, 25 Licenses
Available for managing 25 resources nodes
H3C DR1000 ADDC Fabric Director, 50 Licenses
Available for managing 50 resources nodes
H3C DR1000 ADDC Fabric Director, 100 Licenses
Available for managing 100 resources nodes
H3C DR1000 ADDC Fabric Director, 200 Licenses
Available for managing 200 resources nodes
H3C DR1000 ADDC Fabric Director, 500 Licenses
Available for managing 500 resources nodes
H3C DR1000 ADDC Fabric Director, 1000 Licenses
Available for managing 1000 resources nodes