1. (ITI310) SESSIONS 8: Network Load Balancing (NLB)
By Eng. BASSEM ALSAID

2. “Network Load Balancing (NLB)”
objectives in this session:
Overview.
Features in Network Load Balancing.
Models of Network Load Balancing.
Installing Network Load Balancing.
Create a New Network Load Balancing Cluster.
Add a Host to the Network Load Balancing Cluster.
Single Copy Cluster.
Installation of a Single Copy Cluster under Windows 2008 Server.

3. “Network Load Balancing - Overview”
The Network Load Balancing (NLB) feature in Windows Server 2008 enhances the availability and scalability of critical mission servers. A single computer running Windows Server 2008 provides a limited level of server reliability and scalable performance. However, by combining the resources of two or more computers running one of the products in Windows Server 2008 into a single virtual cluster, NLBC can deliver the reliability and performance that Web servers and other mission-critical servers need. All versions of Windows Server 2008 supports Network Load Balance (NLB) for a group of 2 to a maximum of 32 nodes (We consider each server in a cluster as a node).

4. “Network Load Balancing - Overview”
The diagram above shows two connected Network Load Balancing clusters. The first cluster consists of two hosts and the second cluster consists of four hosts. This is one example of how you can use NLBC.

5. “Network Load Balancing - Overview”
In a Network Load Balancing configuration, a virtual IP address is assigned to each group. When a client invokes a service from the cluster, the request will be translated to one of the real addresses of the cluster servers depending to both: original settings and servers availability.

6. “Features in Network Load Balancing”
NLB includes the following features:
Scalability.
High-availability.
Manageability.
Ease of use.
Features in Network Load Balancing
NLB includes the following features:
Scalability
Scalability is the measure of how well a computer, service, or application can grow to meet increasing performance demands. For NLB clusters, scalability is the ability to incrementally add one or more systems to an existing cluster when the overall load of the cluster exceeds its capabilities. To support scalability, NLB can do the following:
Balance load requests across the NLB cluster for individual TCP/IP services.
Support up to 32 computers in a single cluster.
Balance multiple server load requests (from either the same client or from several clients) across multiple hosts in the cluster.
Support the ability to add hosts to the NLB cluster as the load goes up, without bringing the cluster down.
Support the ability to remove hosts from the cluster when the load goes down.
Enable high performance and low overhead through fully pipelined implementation. Pipelining allows requests to be sent to the NLB cluster without waiting for response to the previously sent one.
High-availability
A highly available system reliably provides an acceptable level of service with minimal downtime. To provide high availability, NLB includes built-in features that can automatically:
Detect and recover from a cluster host that fails or goes offline.
Balance the network load when hosts are added or removed.
Recover and redistribute the workload within ten seconds.
Manageability
NLB provides the following manageability features:
You can manage and configure multiple NLB clusters and the cluster hosts from a single computer by using NLB Manager.
You can specify the load balancing behavior for a single IP port or group of ports by using port management rules.
You can define different port rules for each Web site. If you use the same set of load-balanced servers for multiple applications or Web sites, port rules are based on the destination virtual IP address (using virtual clusters).
You can direct all client requests to a single host by using optional, single-host rules. NLB routes client requests to a particular host that is running specific applications.
You can block undesired network access to certain IP ports.
You can enable Internet Group Management Protocol (IGMP) support on the cluster hosts to control switch flooding (when operating in multicast mode).
You can remotely start, stop, and control NLB actions from any networked computer that is running Windows by using shell commands or scripts.
You can view the Windows event log to check NLB events. NLB logs all actions and cluster changes in the event log.

7. “Models of Network Load Balancing”
Each server in a Windows Server 2008 NLBC could have two network adapters. This is recommended for maximum performance. In this case a first adapter is used for communication between cluster nodes and the other for communication with the clients. Generally, there are four Network Load Balancing models:
Unicast with Single Network Adapter.
Unicast with Multiple Network Adapters.
Multicast with Single Network Adapter.
Multicast with Multiple Network Adapters.
Models of Load Balancing
Each server in a Windows Server 2008 NBLC could have two network adapters. This is recommended for maximum performance. In this case a first adapter is used for communication between cluster nodes and the other for communication with the clients. Generally, there are four Network Load Balancing modes:
Unicast with Single Network Adapter:
One MAC address used as a physical address for the cluster.
Traffic is received by all nodes in the cluster and filtered by the NLB driver.
Nodes in the cluster are able to communicate with addresses outside the cluster
Nodes are not able to communicate with each other.
Unicast with Multiple Network Adapters:
Nodes are able to communicate with each other.
Multicast with Single Network Adapter:
There is a MAC address used as a physical address for the cluster, and other addresses for network adapters.
Nodes in the cluster are able to communicate with addresses outside the cluster.
Multicast with Multiple Network Adapters:
Used for environments where there are significant levels of traffic directed to specific cluster nodes.

8. “Configuring Port”
Network Load Balancing is configured on a port by port basis or by range of ports. There are 3 options for controlling traffic:
Single Host: Traffic is forward to a node in the cluster.
Multiple Hosts: Traffic is distributed between the nodes in the cluster.
Disabled: No Action and no filter.

9. servermanagercmd.exe -install nlb
Practice I:
1. Installing Network Load Balancing.
2. Configuring NLB under Windows Server 2008.
3. Adding Nodes to NLB.
1- Installing Network Load Balancing:
To open the Add Features Wizard and install NLB:
Click Start, click Administrative Tools, and then click Server Manager.
In the Features Summary area of the Server Manager main window, click Add Features.
—or—
In the Customize this server area of the Initial Configuration Tasks window, click Add Features.
In the Add Features Wizard, select the Network Load Balancing check box.
Click Install.
Alternatively, you can install NLB by typing the following command:
servermanagercmd.exe -install nlb
After NLB is installed, you can configure an NLB cluster by using Network Load Balancing (NLB) Manager.
To open NLB Manager, click Start, click Administrative Tools, and then click Network Load Balancing Manager. You can also open Network Load Balancing Manager by typing Nlbmgr at a command prompt.
2- Configuring NLBC under Windows Server 2008
Network Load Balanced clusters are built using the Network Load Balancing Manager:
Go to Start  All Programs  Administrative Tools
Invoke Network Load Balancing Manager
Select Options  Credentials and enter an account and password from the group of administrators (This will help to configure the account and password used on each node in the cluster).
Right click on Network Load Balancing Clusters entry
Select the New Cluster option.
In the dialog New Cluster enter the name or IP address of the first server to be included in the cluster.
Press the Connect to establish a connection with the server.
The server will be listed
Click Next go to the Host Parameters screen.
Set the Priority (unique host ID) which is a number between 1 and 32. It indicates the number of the server within the cluster and its priority.
Set the Dedicated IP addresses used for both communication between cluster nodes and external network traffic.
Click Next in order to display the Cluster IP addresses which are the virtual IP addresses by which the cluster will be accessible on the network.
Click Next to go to the Cluster Parameters screen.
Enter the full internet name of the cluster
Choose the Cluster operation mode.
Click Next to proceed to the Port Rules screen in order to configure ports and to finish with cluster configuration.
3. Adding Nodes to NLB:
To add additional nodes to a Network Balanced Cluster:
Right click on the cluster the Network Load Balancing Manager
Select Add Host To Cluster.
If no cluster is listed, right click on Network Load Balancing Manager entry and select Connect to Existing.
Enter the name or IP address of a node in the cluster.
To add a node, right click on the cluster name and select Add Host to Cluster.
Enter the name or IP address of the host to be added as a new cluster node and click on Connect.
Click Next to change the cluster port rule settings.
The new host is listed in the manager screen with a status of Converged.

10. “Single Copy Cluster”
A single copy cluster (SCC) is a mailbox server cluster that uses shared storage configuration in a failover cluster to allow multiple servers to manage a single copy of storage groups. If the node running a mailbox server cluster has problems, the mailbox server cluster goes offline briefly and let another node takes control of the mailbox server in the cluster. This process is called failover. Administrator can manually move a server mailbox clustered between nodes in a cluster. This process is called a transfer process.
A single copy cluster (SCC) is a mailbox server cluster that uses shared storage configuration in a failover cluster to allow multiple servers to manage a single copy of storage groups.
This feature is similar to cluster features in earlier versions of Microsoft Exchange. Some important changes and enhancements were made. The way you create, manage and troubleshoot a single copy cluster is different. In addition, the failover behavior has changed in Microsoft Exchange Server 2007.
In a single copy cluster (SCC), a mailbox server Exchange 2007 uses its own network identity, not the identity of a node in the cluster. This is known as a mailbox server cluster.
If the node running a mailbox server cluster has problems, the mailbox server cluster goes offline briefly and let another node takes control of the mailbox server in the cluster. This process is called failover.
The storage of the mailbox server cluster is hosted on shared storage accessible to each node can host the mailbox server cluster. When failover occurs, the storage associated with the mailbox server in a cluster is detected in a logical way and placed under the control of the new node's host mailbox server cluster.
However, an administrator can manually move a server mailbox clustered between nodes in a cluster. This process is called a transfer process.

11. “Installation of a Single Copy Cluster under Windows 2008 Server”
Main phases of Single Copy Cluster deployment:
Configuring storage.
Configuring Network.
Training the cluster.
Network configurations clusters.
Validate the failover cluster.
Installation of a Single Copy Cluster under Windows 2008 Server
The installation of a single copy cluster (SCC) on Windows Server 2008 is done in several phases. Although the process of deploying an SCC on Windows Server 2008 deployment is similar to an SCC on Windows Server 2003.
Before performing any of the procedures outlined below, we need to ensure that computers have the specified components of operating system requirements for Windows Server 2008 installed.
It is recommended to complete each phase before starting the next. Once all phases completed, it is recommended to check the SCC solution before using it for production.
The deployment of an SCC on Windows Server 2008 is a process running in several distinct stages. The following sections describe the main phases of cluster deployment:
Configuring storage:
SCC uses a shared storage for a mailbox server cluster. As the SCC using shared storage, all storage must be configured on each node before the configuration of the cluster.
In a SCC, the correct order of installing and configuring storage resources is as follows:
The storage solution must be properly connected and configured at the hardware level before the creation of the failover cluster.
One or more physical disk resources for the mailbox server cluster must exist in the failover cluster before installing Microsoft Exchange Server The installation of Exchange 2007 did not continue if the shared storage is not detected in the cluster.
The dependencies of physical disk resources must be configured manually by an administrator after the addition of a cluster mailbox server cluster.
Configuring Network:
We must have a sufficient number of available IP addresses if we create servers mailboxes clustered in a SCC on Windows Server 2008.
The functionality of failover cluster from Windows Server 2008 includes new network capabilities that constitute a major progress compared to how things were done in clusters inherited. For example, failover cluster Windows Server 2008 introduces support for multiple subnets, the management of Internet protocol DHCP (Dynamic Host Configuration Protocol) version 4 (IPv4) and IPv6.
All IP addresses associated with a Resource Network Name is stored in a dynamic Domain Name System (DNS) (if configured for dynamic updates). As of the cluster nodes can be routed over different networks, the communication mechanisms have been modified to use protocols implemented over UDP (User Datagram Protocol) and based on the implementation of reliable sessions. This will allow network configurations for geographically dispersed clusters will be possible. Therefore, organizations can deploy a cluster failover between data centers without having to physically use LAN (VLAN) technologies to expand the sub-cluster networks between the different locations.
When moving over a mailbox server cluster deployed in a failover cluster subnet geographically dispersed, the name of the mailbox server cluster is preserved but the IP address assigned this name is not. The availability of this server to clients and other servers depends on the spread of the new IP address on the DNS. DNS propagation can take some time. It is therefore recommended to set (TTL, Time to Live) value (for the DNS host record for mailbox server cluster) to 10 minutes.
Training the cluster:
Failover cluster is formed when the first node is added to the cluster. Through this process, the cluster has a network name and IP address of a single network. The network name and IP address, which together form the network cluster identity, are moving between the nodes in the cluster as they are placed online and offline.
Generally, the network identity of the cluster is used rarely in the administration of a mailbox server cluster.
Network configurations clusters:
Once all nodes have been added to the cluster, the cluster components must be configured. Specifically, we must configure the networks of clusters, and the tolerance settings for missed cluster. It is also recommended to rename networks of clusters with more meaningful names.
Validate the failover cluster:
Windows Server 2008 includes new Configuration and Validation Wizard to check the integrity and configuration of a failover cluster. Configuration and Validation Wizard includes four groups of test which are: inventory, network, storage and system configuration.

12. NEXT SESSION
C9: Saturday 28-Jun :00 C10: Saturday 28-Jun :30 Title: “Active Directory – PART 1”
THANKS