IBM System x Private Cloud Offering, Advanced Configuration: Architecture and Implementation Guide

Redpaper
Scott Smith

IBM System x Private Cloud Offering,
Advanced Configuration: Architecture and
Implementation Guide

Introduction
The IBM® System x® Private Cloud Offering provides solutions that are part of the Microsoft
Fast Track for Hyper-V program. This program has requirements for systems management,
virtualized management, and fault tolerance through the use of the System Center
management platform, clustering, and redundancy in the network and storage.

This document provides setup and configuration details for the IBM 8-Node Hyper-V Fast
Track configuration based on the IBM System x3755 M3 servers. Guidance and references
are given throughout this document to aid understanding.

This document is intended for experienced IT engineers who have familiarity with Windows,
Hyper-V, Microsoft Clustering, Networking, and iSCSI storage with IBM System Storage® N
Series products. You should already have hands-on experience configuring these systems.

This paper describes planning and configuration of the solutions. It has the following topics:
“Racking and power distribution” on page 3
“Networking and VLANs” on page 4
“Active Directory” on page 10
“Storage” on page 10
“Setup of IBM System x3550 M3 for Active Directory” on page 14
“Setup of x3650 M3 for Management Cluster” on page 15
“Setup of x3755 M3 for Production Hyper-V Cluster” on page 21
“Conclusion” on page 24
“Appendix A: Reference documents” on page 24
“Appendix B: Brocade 2-port 10GbE Converged Network Adapter” on page 26
“Appendix C: IBM Ethernet Switch B24X” on page 31
“Appendix D: Worksheets” on page 35
“Appendix E: SnapVault sample code” on page 39

© Copyright IBM Corp. 2011. All rights reserved. ibm.com/redbooks 1

Overview of architecture solution
This solution is based on the following clustered infrastructure:
2-node cluster running Hyper-V with virtual machines (VMs) running management tools
8-node cluster running Hyper-V for the production VM environment

This configuration is based on the hardware components in Table 1.

Table 1 Configuration components
Component Quantity

IBM System x3550 M3 1U server (optional Active Directory server) 1

IBM System x3650 M3 2U servers (management servers) 2

IBM System x3755 M3 2U servers (Hyper-V servers) 8

IBM System Storage N6210 storage controller 1

EXN3000 disk expansion trays 5

IBM Ethernet Switch B24X 10GbE switches 2

Brocade 10GbE Dual Port CNA for IBM System x Adapters 20

A diagram of the overall architecture is shown in Figure 1 on page 3.

2 IBM System x Private Cloud Offering, Advanced Configuration: Architecture and Implementation Guide

VLAN 10 iSCSI Storage

VLAN 20 iSCSI Storage

The primary iSCSI
VLAN 30 Production Cluster Private/CSV storage network would
use a pair of dedicated

IBM/Brocade B24X
IBM/Brocade B24X
10GbE links and
VLAN 40 Production Cluster Public/VM Comm MPIO. All remaining
communication is
VLAN 50 Production Cluster Private – Live Migration achieved through the
use of an FT pair of
10GbE links with
VLAN 60 Management separate VLAN
created for each
VLAN 70 Management Cluster Private network needed.

VLAN 80 SQL Cluster Private

IBM N6210

IBM System x3550 M3
2 - 1GbE - Management

2 - IBM System x3650 M3

EXN3000 Expansion
• SQL Server
• System Center Ops Mgr
• System Center VMM
• Self Service Portal
8 - IBM System x3755 M3 • ForeFront Security

Windows Server 2008 R2 1 - FT Pair 10GbE - Storage
Datacenter running Hyper-V • VLAN 10 iSCSI
1 - FT Pair 10GbE - Storage • VLAN 20 iSCSI
• VLAN 10 iSCSI 1 - FT Pair 10GbE - Communication A variety of configuration options
• VLAN 20 iSCSI • VLAN 10 (iSCSI iSCSI Storage) is available to maximize IOPS
1 - FT Pair 10GbE - Communication • VLAN 20 (iSCSI VM Storage) and space: 240 disks maximum,
• VLAN 10 (iSCSI VM Storage) • VLAN 60 (Mgmt Cluster Pub) SnapManager, SnapVault,
• VLAN 20 (iSCSI VM Storage) • VLAN 70 (Mgmt Cluster Priv) SnapMirror for Hyper-V for
• VLAN 30 (Prod Cluster Priv) • VLAN 80 (SQL Cluster Priv) backup to disk.
• VLAN 40 (Prod Cluster Pub)
• VLAN 50 (Prod Cluster LM)
• VLAN 60 (Mgmt Network)

Figure 1 Solution architecture

Racking and power distribution
Installation of the power distribution units (PDUs) and their cabling must be performed before
any of the systems or storage is placed in the rack. There should be sufficient PDUs to allow
redundant power supplies to be fed from separate electrical circuits. Consider the following
information when you cable the PDUs:
Ensure sufficient separate electrical circuits and receptacles to support the PDUs needed.
To minimize the chance of a single electrical circuit failure taking down a device, ensure
that the PDUs, which feed redundant power supplies, are supplied from separate electrical
circuits.
For devices that have redundant power supplies, plan for one electrical cord from separate
PDUs.
Each 10 Gb Ethernet switch should be powered from a separate PDU.
Locate Ethernet switches to allow the easiest cable runs possible.
Locate storage in lower portions of the rack for added stability.

IBM System x Private Cloud Offering, Advanced Configuration: Architecture and Implementation Guide 3

Networking and VLANs
A total of four Converged Network Adapter (CNA) ports will be available on each server
running Hyper-V (two adapters in each server, each with two ports). The networks necessary
to support this configuration are achieved with a combination of physical and virtually isolated
networks.

Eight networks (VLAN IDs 10, 20, 30, 40, 50, 60, 70, and 80) are needed to build this
configuration. All physical servers will use two CNA ports (one from each card) to connect to
the iSCSI storage through a dedicated pair of 10GbE links. Use the IBM N Series
device-specific multipath I/O (MPIO) driver for fault tolerance and load balancing. These
storage connections are made on VLAN 10 and 20. The remaining networks are created by
using the remaining two CNA ports and the Brocade adapter teaming software to create a
fault tolerant (FT) NIC pair from the communications network. VLANs 30, 40, 50, 60, 70, and
80 are created by using the Brocade teaming utility to use this FT team as described in
“Brocade network adapter teaming” on page 6.

For more technical information about the IBM B24X switch, see “Appendix C: IBM Ethernet
Switch B24X” on page 31.

To assist in planning of the networks needed for this configuration see “Appendix D:
Worksheets” on page 35, which lists each VLAN and the servers on each respective network.


VLAN description
The eight VLANs are described in Table 2.

Table 2 VLAN definitions
Network Name Description

VLAN 10 iSCSI Storage Network Used for all iSCSI storage traffic. Physical
servers maintain dedicated physical FT
connection to this network. Virtual machines
use a VLAN created from the
communications network.

VLAN 20 iSCSI Storage Network Used for all iSCSI storage traffic. Physical
servers maintain dedicated physical FT
connection to this network. Virtual machines
use a VLAN created off the communications
network

VLAN 30 Production Cluster Private & CSV Used for cluster heartbeat and Cluster
Network Shared Volume (CSV) traffic

VLAN 40 Production VM Communication Network Used as primary communication network for
production VMs (also known as Cluster
Public Network)

VLAN 50 Production Live Migration Second private network for production
cluster used for live migration.

VLAN 60 Management Network Network used for Active Directory (AD) and
all communication for management of
servers.

VLAN 70 Management Cluster Private Network Used for cluster heartbeat and CSV traffic.

VLAN 80 SQL Cluster Private Network Used for cluster heartbeat traffic.

Switch port layout
This configuration uses two IBM B24X 10 Gb Ethernet switches, each with 24 10GbE ports
and four 1GbE ports. The first twelve ports are for iSCSI communication to the storage
through VLAN 10 and VLAN 20. The best practice for configuring MPIO for iSCSI is to have
separate VLANs for each iSCSI path. In addition, one port from each B24X switch can be
used to connect the two switches. Port recommendations for B24X switches are as follows:
Ports 1 - 2: N6210 Storage (for iSCSI storage network)
– Untagged at the switch on VLAN 10 (Switch1)
– Storage must also be configured to send traffic out tagged on VLAN 10 and 20
Ports 3 - 10: Production nodes (for iSCSI storage network)
Ports 11 - 12: Management nodes (for iSCSI storage network)


Port 13 (for cross-switch traffic)
– Tagged at the switch with all VLANs
Port 14: CorpNet uplink
– Connectivity depends on CorpNet switch: Layer 2 and Layer 3 supported
Ports 15 - 16: Management nodes
– Tagged at the switch for VLANs 10, 20, 60, 70, and 80
– Fault tolerant NIC team with interfaces for each VLAN (10, 20, 60, 70, 80) must be
created on the server using Brocade adapter software. Enable jumbo frames on the
devices before creating the team. See “Appendix B: Brocade 2-port 10GbE Converged
Network Adapter” on page 26.
Ports 17 - 24: Production nodes
– Tagged at the switch on VLANs 10, 20, 30, 40, 50, and 60
– Fault tolerant NIC team with interfaces for each VLAN (10, 20, 30, 40, 50, 60) must be
created on the server by using Brocade adapter software. Enable jumbo frames on the
devices before creating the team. See “Appendix B: Brocade 2-port 10GbE Converged
Network Adapter” on page 26.
Port 25: 1 GbE RJ-45 port – Optional AD server
– Untagged at the switch on VLAN 60
Port 26: 1 GbE RJ-45 port – N6210 management port
– Untagged at the switch on VLAN 60
Switch ports must allow the appropriate VLAN traffic as needed.

Figure 2 shows the switch port layout for B24X switch.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 N6210 Mgmt
AD Svr

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 N6210 Mgmt
AD Svr
iSCSI Storage Cross SW Traffic CorpNet Uplink Data Comm

Figure 2 Switch port layout for B24X

Brocade network adapter teaming
Using the Brocade teaming software, the two CNA ports are combined as a fault tolerant
team. This way, in combination with each port going to a separate switch, can provide fault
tolerance for this configuration. In addition, the teaming software can be used to create
individual VLAN devices off this team.

Before the team is created, both devices must be set to support 9000 byte jumbo frames
under device manager. The VLANs created from these devices can then support jumbo
frames if needed. The VLANs supporting iSCSI to a VM would be using jumbo frames from
the teamed devices.


Figure 3 shows the creation of fault tolerant NIC team, and the associated MAC address for
the team. Multiple VLANs can then be created from this team as shown.

Figure 3 Brocade teaming software and VLAN creation

These devices are listed under Windows “Network Connections” and can be assigned an
appropriate IP address, or visualized under Hyper-V for use by virtual machines. Consider
setting a Windows quality of service (QoS) limit to bandwidth to an appropriate level for
each network. See the Hyper-V: Live Migration Configuration Guide:
http://technet.microsoft.com/en-us/library/ff428137(WS.10).aspx

The networks are each described in the following sections.

iSCSI storage networks (VLANs 10 and 20)
These networks are reserved for server access to the iSCSI storage network. All iSCSI traffic
will be isolated on VLAN 10 and VLAN 20.

Storage controller access
Note the following information:
There should be two Ethernet connections from each storage controller to the IBM B24X
Ethernet switches. Each storage controller should maintain a 10 Gb Ethernet connection
to VLAN 10 and VLAN 20. This way provides redundancy if one link fails.
Set the switch and storage controller to support 9000 byte jumbo frames.


Physical host storage access
Each physical server must maintain two 10 Gb Ethernet connections to these storage
networks, one connection each to VLAN 10 and VLAN 20. One port on each CNA is used
to provide fault tolerance and terminate in the assigned port in each physical switch for
redundancy.
Ten ports on each of the two physical switches must be reserved for server connections to
the dedicated iSCSI networks.
Set the switch ports and CNA ports to use 9000 byte jumbo frames.
Connect the following devices to the respective VLANs at each switch:
– Each of the two management servers (two ports)
– Each of the eight production Hyper-V servers (eight ports)
The N6210 MPIO device-specific module must be installed and used to manage the two
connections for fault tolerance and load balancing.
Load balancing can be achieved by setting the properties of the device to Round Robin
from within the Server Manager as shown in Figure 4.

Figure 4 Setting MPIO load balancing policy

Virtual machine storage access
Virtual machines might require access to storage for either guest clustering or connection
to pass-thru disks. These connections can be achieved in one of several ways:
– A pass-thru disk is exposed to the parent partition and seen under disk manager of the
physical machine. If left offline in the parent machine, it can then be made available to
a virtual machine under the disk settings of the VM.
– For direct connections to storage, such as for configuring guest clustering in virtual
machines, the Microsoft iSCSI Initiator is available under Windows 2008 R2 and can be
used to map available LUNs from the storage controller to specific virtual machines
directly.
To support this approach, the VLAN 10 and VLAN 20 network connections must be
made available to these virtual machines. These VLANs must be created under the
NIC team that is created on the data communication network, with the remaining to the
CNA ports as seen in “Brocade network adapter teaming” on page 6.
The IBM N Series DSM MPIO driver must also be loaded in the virtual machine.
The Brocade adapter software should tag this network device, and the switch be
configured to allow VLAN 10 and VLAN 20 traffic from this port to communicate on the
iSCSI storage networks.


At a minimum, the two physical management servers require virtual machine storage
access to support the guest clustering of two virtual machine SQL servers.
Consider making QoS settings on VM storage access to cap I/O traffic and not overwhelm
network communication for the remaining virtual machines.

Cluster Heartbeat and Cluster Shared Volume Networks (VLANs 30, 70, and 80)
These networks are reserved for cluster private (heartbeat) communication between
clustered servers. The management servers, Hyper-V production servers, and SQL servers
have their own virtually isolated networks, created by using the Brocade adapter software to
create separate VLANs. Switch ports must be configured to limit the scope of each of these
VLANs as appropriate. There should be no routing on any VLAN that is being used for cluster
private networks.

The following devices should access this network:
Each of the two management servers: VLAN 70 should be unique for this cluster
Each of the eight production Hyper-V servers: VLAN 30 should be unique for this cluster
Each of the two SQL Server virtual machines: VLAN 80 should be unique for this cluster

Network interfaces for each of these VLANs are configured on the server through the use of
the Brocade adapter teaming software.

Production Virtual Machine Communication Network (VLAN 40)
This network can support virtual machine communication for the Hyper-V production cluster.
VLAN 40 is created from the FT NIC team as described in“Brocade network adapter teaming”
on page 6. The production Hyper-V servers use VLAN 40 to create a virtual NIC through
Hyper-V to be presented to production virtual machines. The production servers themselves
do not have to maintain an IP address on this network.

Production Live Migration Network (VLAN 50)
Create VLAN 50 to support Live Migration for the production cluster. Create VLAN 50 as part
of the FT team supporting the communication network using the Brocade adapter teaming
software. Consider setting a QoS limit (see “Appendix B: Brocade 2-port 10GbE Converged
Network Adapter” on page 26) to ensure plenty of bandwidth remains for VM communication
during live migration.

Management Network (VLAN 60)
This network is reserved for management of the Hyper-V cloud. The System Center
management servers communicate with the Hyper-V servers and supporting devices across
this network. The two clusters of physical machines use their respective connections to
support this communication, as needed.

The following devices and servers are on this network (VLAN 60):
The N6210 storage controllers
Each of the two physical management servers
Management Virtual Machines (SCOM, SCVMM, SQL, SSP)


Supporting these virtual machines requires the creation of virtual NICs from this device
under Hyper-V for presentation to the management VMs
Each of the eight production Hyper-V servers
This network is reserved for management and must not be exposed to the VMs

Routing summary
Configure routing on the switches to allow traffic between VLAN 50 and VLAN 60, and other
corporate network resources as needed. Do not enable routing for VLAN 10, 20, 30, 40, 70,
and 80 because these are private networks that support the operations of this configuration.

Active Directory
The Hyper-V Fast Track configuration must be part of an Active Directory domain. This way is
required to form the clusters, and by some of the System Center management components. A
domain controller is included in this configuration as shown in Figure 1 on page 3, but may be
replaced by an existing domain structure if available.

A total of three distinct environments are created:
An eight node cluster of Hyper-V production servers to host virtual machines to various
users
A two node Hyper-V cluster to manage the virtual environment
A pair of clustered SQL virtual machines to provide DB services to the management
cluster

Develop a consistent naming convention for the physical and virtual servers in this
environment. For example the production Hyper-V servers might be named Hyper1, Hyper2,
through Hyper8).

Storage
This configuration uses the IBM System Storage N6210 storage controller with optional
10GbE ports as the primary storage for the management and Hyper-V production clusters.
The configuration that is used for this reference architecture consists of N6210 controller and
five EXN 3000 expansion trays for a total of 120 disks. These disks will be configured and
distributed across both storage controllers to balance the I/O workload

The storage subsystem used by this configuration will need to support the 2-node
management cluster, and the 8-node Hyper-V production cluster. Storage I/O can vary widely
depending on the applications being run. Therefore, be sure to profile your storage needs to
ensure adequate space and I/O capacity for your needs. The storage layout shown in
Figure 5 on page 11 provides a list of disk volumes needed on the N Series storage to
support this configuration.


Storage: IBM N6210
Volumes/LUNS

Cluster Witness Disks:
Hyper-V Production Cluster (1GB)
Management Cluster (1GB)
SQL Cluster (1GB)

Production CSV Disks:
Volume1 – VM OS/Config Files (2.5TB)
Volume2 – VM OS/Config Files (2.5TB)
Volume3 – Database/Random I/O (2.5TB)
Volume4 – Logging/Sequential I/O (2.5TB)

Management CSV Disks:
Volume1 – VM OS/Config Files (500GB)
Volume2 – VM OS/Config Files (500GB)

Additional Production Disks:
SnapManagerHyperV (20GB)
Others as needed Up to 10 – EXN3000 for a
Additional Management Disks: total of 240 disks per N6210
SCVMMLib (500GB – PassThru)
SQLData (50GB) Potential 120 Disk Configuration
SQLLog (20GB) 8 Disks Aggregate1 – Controller-A
8 Disks Aggregate2 – Controller-B
Optional SnapVault Disks:
One volume for each volume to 51 Disks Aggregate1 – Controller-A
backup 51 Disks Aggregate2 – Controller-B

2 Disks for Hotspare – 1 per Controller

Figure 5 Storage layout

Additional storage I/O cards
The IBM N6210 has the following I/O cards installed to support operations:
10GbE PCIe cards installed (one card per controller for a total of two fault-tolerant 10GbE
ports)
Two 256GB Flash Cache Modules (one per controller)

Cabling
The IBM N6210 must be powered from two separate PDUs operating from unique electrical
circuits to provide maximum fault tolerance. Each EXN3000 disk expansion drawer must be
connected to PDUs in the same fashion as the N6210.

Each EXN3000 expansion drawer must be connected to the N6210 over a SAS connection.
To maximize fault tolerance and performance, see the N Series documentation for cabling
directions.

Each 10 Gb Ethernet connection must be plugged into its assigned port on the IBM Brocade
B24X switch:
Configure these ports as untagged ports on the switch in VLAN 10 and VLAN 20.
Enable the 9000 byte jumbo frames on the controller, switch, and server adapters.


One connection from each 1GbE management port on the N6210 should be connected into
the assigned 1GbE ports on each B24X switch. These ports are configured as untagged
ports and assigned to VLAN 60 on the switch.

Management
Management of this storage is performed using a web-based management tool for the
following actions:
1. Controller setup and license enablement
2. RAID Group configuration
3. Aggregate creation
4. Volume creation
5. LUN creation
6. iSCSI target assignments

To begin management of the IBM N6210, complete the following actions:
1. Verify one connection from each controller is connected to a 1 GbE RJ45 port on each of
the B24X switches.
Set the 1 GbE management connections for each controller to the assigned IP address on
VLAN 60.
2. If needed establish a serial connection to the controller (or controllers) to set the IP
addresses for each management connection.
3. Ensure each of the 10GbE connections from each controller are connected to the
appropriate ports on the switches.
4. Using a browser, connect to the N6210 storage unit and set the appropriate IP addresses
for each controller on each of the two 10GbE ports:
– The address is as follows:
http://N6210-IP-address/na_admin
– The default user name is root, and no password (change the root password to
something secure).
5. Enable the required licenses for the following features:
– SnapDrive (storage management tool)
– Snap Manager for Hyper-V (snapshot tool for Hyper-V)
– SnapRestore (snap restore tool)
– SnapVault (optional, if backing up to non-production storage)

Configuration
Before you configure the disks, note the following information:
Verify all disks are available and online before you begin.
Leave one disk available as a hot spare for each controller.


To configure the disks, use the following steps:

Note: With the numbers listed, the assumption is that 120 - 600 GB 15K SAS disks are
available to the storage controllers.

1. Create RAID Groups that use RAID double parity (DP).
This enhanced RAID 6 (double parity) is optimized for the IBM N Series storage:
– RG1 = 8 Disks RAID DP
2. Create aggregates from the RAID Groups in the previous step.
Aggregates consist of multiple RAID groups and allow storage pools that consist of many
more disks than is optimal for a single RAID group. I/O can then be spread across a
greater number of spindles, increasing the performance of the storage system.
Create the following aggregates:
– AG1 = RG1 (approximately 1.7 TB)
• This aggregate is reserved for volumes that support management cluster.
• Assign Controller-A as the preferred controller.
– AG2 = RG2 (approximately 1.7 TB)
• This aggregate is reserved for volumes supporting management cluster.
• Assign Controller-B as the preferred controller.
– AG3 = RG3 + RG4 + RG5 (approximately 17.4 TB)
• This aggregate is reserved for volumes supporting Hyper-V production cluster.
• Assign Controller-A as the preferred controller.
– AG4 = RG6 + RG7 + RG8 (Approximately 17.4 TB)
• This aggregate is reserved for volumes supporting Hyper-V production cluster.
• Assign Controller-B as the preferred controller.
3. Create volumes.
Create the volume to support operations while distributing the I/O load across both
controllers. Volumes are created from aggregates, and LUNS are created from volumes.
Create one LUN from each volume. When volumes are created there is typically a 20%
reserve set aside for snapshots. Sizes given here are after snapshot reserves are
accounted for.


The following volume configuration can be used to support this reference configuration:
– From AG1 and AG2, the following volumes can be created and masked to the
management cluster physical machines. The SQL cluster quorum will be masked to
the IQNs of the SQL VMs, when known:
• 500 GB management cluster CSV LUN
• 500 GB SCVMM library LUN
• 1 GB Management Cluster Quorum LUN
• 1 GB SQL cluster quorum LUN
• SQL DB disk (50 GB size varies by environment)
• SQL Log Disk (20 GB size varies by environment)
– From AG3 and AG4, the following volumes can be created and masked to the Hyper-V
production servers participating in the eight-node cluster:
• Four 2.5 TB production CSVs for VM configuration and OS files
• 20 GB production SnapInfo LUN
This LUN can be created using the SnapDrive utility as a clustered resource after
the cluster has been created.
• 1 GB production cluster quorum LUN
• Optional staging LUN
– If SnapVault will be used to copy disk images from production disks, additional space is
required:
• SnapVault requires enough space to make the initial copy of the data, plus the
incremental updates going forward
• Volumes must be created as part of a qTree. A qTree-based volume provides
additional flexibility and supports an extended setup of commands.
• One volume is necessary for each volume to be backed up.

Multipath I/O (MPIO) fault-tolerance driver
The IBM N6210 provides fault tolerance with redundant controllers. It uses a device-specific
MPIO driver to provide balanced I/O across the two redundant controllers. The driver to
support this approach will be loaded in the physical servers making connections to the
N6210. You may download this driver from the IBM N Series support site after properly
registering:
https://www.ibm.com/support/entdocview.wss?uid=ssg1S7003467

An example of configuring the servers can be seen in Figure 4 on page 8.

Setup of IBM System x3550 M3 for Active Directory
An Active Directory (AD) server is required so that the servers can participate in a Microsoft
cluster and is needed by several parts of the management solution. If an AD server is already
available in your environment, the x3550 M3 offered as the AD server in this configuration is
not needed.

The setup of the IBM x3550 M3 is straightforward because no optional devices are installed in
this system.


Take the following actions before software configuration:
1. Confirm one onboard NIC port is connected to one of the 1 GbE RJ45 ports on each of the
B24X switches.
2. Confirm that the B24X switch port is configured for VLAN 60.

Be sure the two local disks are configured as a RAID 1 set. For more information, see
ServeRAID M1015 SAS/SATA Controller User's Guide:
ftp://ftp.software.ibm.com/systems/support/system_x_pdf/ibm_doc_sraidmr_m1015_2nde
dition_userguide.pdf

See “Appendix A: Reference documents” on page 24 for links to other resources.

Next, complete the following actions:
1. Install Windows Server 2008 R2 SP1 Standard Edition.
2. Assign fixed TCP/IP address (VLAN 60).
3. Verify network connectivity.
4. Ensure the system UEFI code and integrated management module are updated to the
latest supported version and latest device driver. The x3550 M3 driver matrix are located
at the following address:
http://www-933.ibm.com/support/fixcentral/systemx/selectFixes?parent=ibm~System
x3550M3&product=ibm/systemx/7944&&platform=Windows+2008+x64&function=all
5. Promote to domain controller.
6. Run Windows Update.

System service accounts are added to the domain to support the System Center components
that will be added later. At a minimum, the following accounts should be created:
SVC_SCOM
SVC_SCVMM
SVC_VMMSSP
SVC_SQL

Setup of x3650 M3 for Management Cluster
The management cluster consists of two IBM x3650 M3 systems with 24 GB of RAM each.

Complete the following actions before software configuration:
1. Confirm that two dual port CNA cards are installed.
2. Confirm one port from each CNA card is connected to the assigned storage network port
(VLAN 10 and VLAN 20) on each respective B24X switch.
3. Confirm one port from each CNA card is connected to the communications network ports
(VLAN 60, 70, and 80) on each B24X switch.

Set up involves installing Windows Server 2008 R2 Datacenter Edition on each server,
validating networking, and storage connectivity, and enabling Hyper-V and Microsoft
Clustering. The Datacenter version of the Windows product is preferable because of the
unlimited number of Windows Server 2008 R2 virtual machines that can be run under the
Datacenter without further licensing. The cost of the Datacenter version is based on the
number of populated sockets on the server.


Configure the two local disks as a RAID 1 array. For more information, see the ServeRAID
M1015 SAS/SATA Controller User's Guide:

“Appendix A: Reference documents” on page 24 has links to additional resources.

1. Install and update Windows Server 2008 R2 SP1 Datacenter Edition.
2. Join Domain.
3. Enable Hyper-V.
When you create VMs, consider changing the VM memory allocation from static to
dynamic to allow greater flexibility of system resources
4. Enable Windows failover clustering.
5. Run Windows Update
6. Ensure that the system UEFI code and integrated management module are updated to the
latest supported version and latest device drivers. The x3650 M3 driver matrix is at the
following web address:
7. Install N Series device-specific MPIO driver.
The N Series drivers and updates are available from your registered support link with IBM.
8. Install N Series Appliance Watch Pro driver.
9. Install N Series SnapDrive utility.
10.Install N Series SnapManager for Hyper-V.
11.IBM Director Platform agent for System Center upward integration.

Network configuration
In this section, we cover network configuration.

Physical server storage network
Complete the following steps to configure the network:
1. Assign fixed IP addresses to the two CNA connections that are attached to the iSCSI
storage network. Use one port from each CNA; each port should be connected to a
separate B24X switch.
2. Assign IP address for each device on VLAN10 & VLAN20 ID, respectively.
3. Validate connectivity for each device.


Communication networks
Complete the following steps to configure the network:
1. Using the Brocade teaming software in the parent partition, create a fault tolerant (FT)
team from the remaining two CNA ports that are connected to the communications
network.
2. Using the Brocade teaming software, create VLAN 10 and VLAN 20 interfaces over the FT
team to support guest clustering:
– Be sure no IP address for this device are on the physical servers because this address
will be passed for VM direct connections to the storage only. To ensure this action does
not happen, clear the management access check box when the virtual switch is
created.
– Create Hyper-V virtual switches to be passed to the SQL VMs to support clustering.
3. Using the Brocade teaming software, create VLAN 60 interface over the FT team to
support management communications:
– Assign fixed IP address to each physical server.
– Create a Hyper-V virtual switch to be passed to VMs to support management
operations.
4. Using the Brocade teaming software, create VLAN 70 interface over the FT team as a
cluster private network for use by the two management servers:
– Assign a fixed IP address to the physical servers
5. Using the Brocade teaming software. create VLAN 80 interface over the FT team for use
as cluster private network for the two clustered SQL Servers:
– There should be no IP address for this device on the physical servers.
– Create a Hyper-V virtual switch to be passed to the SQL VMs to support guest
clustering.
6. Using the PING command, validate network connectivity between the respective VLANs
for the physical management servers on VLAN 60 and VLAN 70 for use in the
management cluster creation.

Storage configuration
To ensure that the storage volumes are isolated to the specific servers assigned to them in
the network, unique iSCSI qualified names (IQNs) are used to mask the management
server’s storage. To view this setting, select Control Panel  iSCSI Initiator 
Configuration tab, as shown in Figure 6 on page 18.


Figure 6 iSCSI Server IQN

Complete the following steps to configure the storage:
1. Record IQNs of each management server, being careful to ensure that the server has
already joined the domain because not already joining the domain might change the IQN.
2. Use the IBM N6210 management console to add the recorded IQNs to storage volumes
reserved for the management cluster:
– Management Cluster CSV volume
– Management Cluster Quorum volume
– SCVMM library volume
3. Use the Windows iSCSI Initiator to discover the N6210 storage. A connection should exist
from each CNA to each controller, for a total of four connections.
4. Use Disk Manager to validate that storage on N6210 is visible. Create partitions, and
format the management cluster quorum and any management cluster CSV disks.

Cluster creation
To provide a highly available management environment, a cluster is created between the two
management servers. The management servers will then run highly available virtual
machines (HA VMs) on this cluster. The SQL Server virtual machines will not be HA VMs, but
will form their own cluster to support SQL Server in a highly available environment. If you are
not familiar with setting up a Microsoft cluster under Windows 2008, see “Appendix A:
Reference documents” on page 24, which has links to more information.

First, use the Failover Clustering management console to complete the following steps:
1. Select the validate configuration option to validate the two physical management servers
as a potential cluster, and address any errors found.
2. Create a cluster with the two physical management servers.

1. Validate the cluster name and IP address.
2. Validate cluster networking.
3. Validate storage in cluster.
4. Enable Cluster Shared Volumes and add the designated volume as a CSV.
5. Set default paths for VM creation to use the CSV.


SQL Server VM
Several databases that are necessary to support the System Center servers (SCOM,
SCVMM, SSP) will be hosted on a pair of clustered SQL Server virtual machines. To provide
a fault tolerant environment a two-node SQL Server cluster is created by using guest
clustering in virtual machines, one on each of the two management servers with Hyper-V
installed. Although these virtual machines already reside on a pair of clustered servers, they
must not be enabled as HA virtual machines. Running guest clustering in a HA virtual
machine is not supported.

The following items are required for this configuration:
2 non-HA virtual machines (one on each management server)
4 virtual CPUs
6 GB of RAM
4 Virtual NICs: 1 client connection (VLAN 60), 1 cluster private (VLAN 80), 2 iSCSI (VLAN
10 and VLAN20). The N Series MPIO driver must be installed in the VM for storage
access
Windows Server 2008 R2 Enterprise Edition or Datacenter Edition installed and updated
in the VMs. This version must be a clusterable version of Windows Server 2008 R2.

Storage being used for the SQL Server cluster must be allocated from the available storage in
the management server RAID Group on the N6210, but the volumes must be isolated to the
two SQL Server VMs by using IQN restrictions on the volume. Allocate the following storage:
Operating system virtual hard disk (OS VHD) on CSV volume (30 GB)
LUN1 – SQL databases
LUN2 – SQL logging
LUN3 – SQL cluster quorum (1 GB)

The SQL LUNs must be masked to the two virtual machines.

Use the SQL Service account created on the AD server when you install SQL Server.
Guidance for installing and creating clustered SQL Server 2008 SP1 systems is at the
following address:
http://technet.microsoft.com/en-us/library/ms179530(SQL.100).aspx

The following databases are to be used by System Center:
VMM <VMM_DB>
VMM SSP (Self Service Portal) <SCVMMSSP>
SCOM <Ops_Mgr_DB>
SCOM <Ops_Mgr_DW_DB>
– SCOM has a DB creation utility on the installation media to run on the SQL Servers to
create the required databases.
– SCVMM and SSP create the DB on the SQL Server during setup.

System Center Operations Manager
A System Center Operations Manager server must be available to support the management
environment used in this configuration. To provide fault tolerance this server will be a highly
available virtual machine.


This server is implemented as a virtual machine with the following specifications:
HA virtual machine
2 virtual CPUs
4 GB of RAM
1 virtual NIC – Management Network Connection (VLAN 60)
1 OS VHD

To install System Center Operations Manager complete the following steps:
1. Install and update Windows Server 2008 R2 on VM.
2. Assign fixed IP address on VLAN 60.
3. Install SCOM 2007 R2.
4. Update SCOM.

The following roles are required by Operations Manager:
Root Management Server
Reporting Server (DB will reside on SQL Server)
Data Warehouse (DB will reside on SQL Server)
Operator Console
Command Shell

The following Operations Management packs must be installed:
Virtual Machine Manager 2008 R2 pack
Windows Server Base Operating System pack
Windows Failover Clustering pack
Windows Server 2008 Hyper-V pack
Microsoft SQL Server pack
Microsoft Windows Server Internet Information Server pack
System Center packs
IBM Director Upward Integration pack (see “Appendix A: Reference documents” on
page 24 for links to more information)
IBM N Series integration pack

System Center Virtual Machine Manager
A System Center Virtual Machine Manager server must be available to support the
management environment used in this configuration. To provide a fault tolerance, this server
will be a highly available virtual machine.

The server is implemented as a virtual machine with the following specifications:
HA virtual machine
2 virtual CPUs
4 GB of RAM
1 virtual NIC: Management Network Connection (VLAN 60)
1 OS VHD
1 data VHD or pass-thru disk for VMM Library


To install System Center Operations Manager complete the following steps:
1. Install Windows Server 2008 R2 on VM.
3. Update Windows.
4. Install SCVMM 2008 R2.
5. Update SCVMM.

The following roles are required by Virtual Machine Manager:
VMM Server
Administrative Console
VMM Library
SQL DB will reside on remote SQL cluster
Command Shell

In addition, VMM will be integrated with Operations Manager. VMM uses System Center
Operations Manager to monitor the health and availability of the virtual machines and virtual
machine hosts that VMM is managing (see “Appendix A: Reference documents” on page 24
for links to integration guidance).

System Center Self Service Portal 2.0
System Center Self Service Portal 2.0 must be available to support the management
environment and self-service provisioning of virtual machines. To provide a fault tolerance this
server will be a highly available virtual machine. Microsoft installation and configuration
instructions are available from the following web address:
http://www.microsoft.com/downloads/en/details.aspx?FamilyID=fef38539-ae5a-462b-b1c
9-9a02238bb8a7&displaylang=en#SystemRequirements

The server is implemented as a virtual machine with the following specifications:
HA virtual machine
2 virtual CPUs
4 GB of RAM
1 virtual NIC – Management Network Connection (VLAN 60)
1 OS VHD

To install System Center Self Service Portal 2.0, complete the following steps:
1. Install and update Windows Server 2008 R2 on VM.
3. Install and update IIS with required components.
4. Install and update SSP 2.0.

Setup of x3755 M3 for Production Hyper-V Cluster
The production cluster will consist of eight IBM x3755 M3 systems with 256GB of RAM
each.Before configuring the software, complete the following steps:
1. Confirm that two dual port CNA cards are installed.
2. Confirm one port from each CNA card is connected to the assigned storage network port
(VLAN 10 and VLAN20) on each B24X switch.
3. Confirm one port from each CNA card is connected to the assigned communications
network port (VLAN 30 through VLAN 80) on each B24X switch.


Software installation
Setup involves installing Windows Server 2008 R2 SP1 Datacenter Edition on all servers,
validating networking, setting up storage, and enabling Hyper-V and Microsoft Clustering.

First, the two local disks must be configured as a RAID 1 array. See the ServeRAID M1015
SAS/SATA Controller User's Guide:

“Appendix A: Reference documents” on page 24 has links to additional resources.

Next, complete the following steps:
1. Install and update Windows Server 2008 R2 Datacenter Edition.
2. Join Domain.
3. Enable Hyper-V.
4. Enable Windows Failover Clustering.
5. Ensure the system UEFI code, integrated management module, and latest device drivers
are updated to the latest supported version. Go to the following web address:
6. Install the latest Windows updates.
7. Install IBM N Series device-specific MPIO driver.
8. Install IBM N Series Appliance Watch Pro driver.
9. Install IBM N Series SnapDrive utility.
10.Install IBM N Series SnapManager for Hyper-V.
11.Install Data ONTAP PowerShell Toolkit.
12.Install IBM Director level 1 agent for System Center upward integration.

Network configuration
Assign fixed VLAN 10 and 20 IP addresses to the two CNA connections that are attached to
the iSCSI storage network. Use one port from each CNA. Connect each port to a separate
B24X switch. Note the following information:
No NIC teaming or VLAN assignment is required on the server for these two ports.
Validate connectivity for each device.

Using the Brocade teaming software, create FT team from two remaining CNA ports that are
connected to communications network, as follows:
1. Create a VLAN 10 interface from the FT team if iSCSI access is needed for any virtual
machines on the production network.
A Hyper-V switch should be created for VM use only. Do not allow the physical server
access.
2. Create a VLAN 20 interface from the FT team for the cluster private network.
A Hyper-V switch should be created for VM use only. The physical server should not be
allowed access.


3. Create a VLAN 30 interface from the FT team for the cluster private network.
Assign the VLAN a static IP address, however no Hyper-V switch should be created on top
of this because it is reserved for clustering of the physical machines.
4. Create a VLAN 40 interface on the FT team as a second cluster private network for use by
Live Migration
of this because it is reserved for Live Migration of virtual machines.
Consider setting a Windows QoS limit to bandwidth to an appropriate level for each
network. See the Hyper-V: Live Migration Configuration Guide:
5. Create a VLAN 50 interface on the FT team to support virtual machine communications.
A Hyper-V switch should be created for VM use only. The physical server should not be
allowed access.
6. Create a VLAN 60 interface on the FT team to support management communications.
of this because it is reserved for management purposes.
7. Validate the network connectivity between physical servers on VLAN 30, 40, and 60
respectively.

Storage configuration
To properly enable the storage to be isolated from other servers in the network the unique
IQNs are used to mask the management server’s storage. To view this information, select
Control Panel  iSCSI Initiator  Configuration tab.

Complete the following steps:
1. Record IQNs of each host server, carefully ensuring that the server has already joined the
domain, because the process of joining the domain could change the IQN.
2. Use the IBM N6210 management console to add the recorded IQNs to storage volumes
that are reserved for the production cluster:
– Production Cluster CSV volumes
– Production Cluster Quorum volume
3. Delay the creation of the SnapInfo volume until you have the SnapDrive utility installed and
operational. This step makes the creation and masking of the this volume much easier.
4. Use the Windows iSCSI Initiator to discover the N6210 storage
5. Use Disk Manager to validate that storage on N6210 is visible.
6. Create partitions and format the production cluster quorum, and future production cluster
CSV disks.

If configuring SnapVault to back up snap images to non-production disks, see “Appendix E:
SnapVault sample code” on page 39, which shows a short PowerShell script that can be set
to run as part of the SnapManager for Hyper-V task. Set it to run after the snaps are taken.


Cluster creation
To provide a highly available virtual machine environment, a cluster is made up of the eight
production servers. The servers created from these clustered resources are then run as
highly available virtual machines on this cluster. If you are not familiar with setting up a
Microsoft cluster under Windows Server 2008, see the “Appendix A: Reference documents”
on page 24 for links to more information.

First, use the Failover Clustering management console to complete the following steps:
1. Validate the eight physical production servers as a potential cluster, and address any
errors found.
2. Create a cluster with the eight physical productions servers.

1. Validate cluster name and IP address.
2. Validate cluster networking.
3. Validate storage in cluster.
4. Enable Cluster Shared Volumes and add the designated volumes as a CSV.
5. Set default paths for VM creation to use the Cluster Shared Volume.

Conclusion
After you complete these steps, you have two operational Hyper-V clusters, one for
management, and one for production virtual machines. Fault tolerance is achieved at several
points through the use of clustering, redundant network switches, and redundant storage
controllers for the iSCSI storage.

System Center Operations Manager, with the help of the IBM upward integration module,
monitors the health of the physical system; System Center Virtual Machine manager monitors
the health of the virtual environment and provides the VM deployment mechanism to the Self
Service portal.

System Center Self Service Portal can be used to allocate a subset of the resources available
to this environment to one or more organizations. These organizations can then start or stop
virtual machines as needed. Billing and metering information for the resources used can be
collected and presented to assist in breaking down, and presenting costs to the organizations
that are using these resources.

Appendix A: Reference documents
The following resources provide additional information regarding setup and configuration.
IBM System x3550 M3 Installation and User's Guide
http://ibm.com/support/entry/portal/docdisplay?lndocid=MIGR-5083597


IBM firmware update and best practices:
ftp://ftp.software.ibm.com/systems/support/system_x_pdf/firmware_update_best_pr
actices.pdf
IBM ServeRAID M1015 SAS/SATA Controller guide
ftp://ftp.software.ibm.com/systems/support/system_x_pdf/ibm_doc_sraidmr_m1015_2
ndedition_userguide.pdf
IBM ServeRAID M5014/M5015 SAS/SATA Controllers guide
ftp://ftp.software.ibm.com/systems/support/system_x_pdf/serveraid-m5014_m5015_u
g_1st_edition_20090924.pdf
IBM IMM and uEFI white paper
ftp://ftp.software.ibm.com/systems/support/system_x/transitioning_to_uefi_and_i
mm.doc
Hyper-V and Microsoft clustering
http://technet.microsoft.com/en-us/library/cc732181(WS.10).aspx
Hyper-V Live Migration and QoS configuration
IBM Director Agent download (Platform Agent)
http://ibm.com/systems/software/director/downloads/agents.html
IBM Director System Center integration (download integration modules)
http://ibm.com/systems/software/director/downloads/integration.html
Microsoft Virtual Machine Manager Self Service Portal 2.0
http://www.microsoft.com/downloads/en/details.aspx?FamilyID=fef38539-ae5a-462b-
b1c9-9a02238bb8a7&displaylang=en#SystemRequirements
NetApp SnapManager 1.0 for Hyper-V Best Practices
http://media.netapp.com/documents/tr-3805.pdf
Microsoft System Center Virtual Machine Manager Deployment Guide
http://technet.microsoft.com/en-us/library/ee441285.aspx
Configuring Operations Manager Integration with VMM
http://technet.microsoft.com/en-us/library/cc956099.aspx
Microsoft SQL Server 2008 documentation
http://technet.microsoft.com/en-us/library/bb418470(SQL.10).aspx
Microsoft ForeFront documentation
http://technet.microsoft.com/en-us/library/ff630950.aspx
IBM white paper about SCOM integration
ftp://ftp.software.ibm.com/systems/support/system_x_pdf/ibm_hw_mp_mgt_guide.pdf
IBM website for Microsoft Systems Management integration
http://ibm.com/support/entry/portal/docdisplay?lndocid=SYST-MANAGE


Appendix B: Brocade 2-port 10GbE Converged Network
Adapter
The Brocade 2-port 10GbE Converged Network Adapter for IBM System x is a two-port
Converged Network Adapter (CNA) PCIe 2.0 x8 card that mounts inside supported IBM
System x servers. This card combines the functions of a 10GbE network interface card (NIC)
and Fibre Channel host bus adapter (HBA).

The CNA supports full Fibre Channel over Ethernet (FCoE) protocol offload and allows
Ethernet and storage traffic to run simultaneously over a converged link. Advanced
capabilities such as support for jumbo frames, VLAN tagging, TCP segmentation offload
(TSO) and large send offload (LSO) also helps in iSCSI storage environments.

Note: Be sure you are running driver v2.3 or higher to support the configurations required
for Fast Track.

Features
The Converged Network Adapter supports the following features:
PCI Express x8 Gen 2 host interface
IPv4 and IPv6
Brocade Host Connectivity Manager (HCM) device management and Brocade
Command-Line Utility (BCU) tools
Unified management with IBM System Storage Data Center Fabric Manager (DCFM) and
IBM Systems Director

The adapter has the following features that are related to Ethernet:
10 Gbps throughput per port full duplex
1.8 million packets per second per port in each direction (700 byte packets, latency < 2 µs)
Checksum/CRC offloads for FCoE packets, IPv4/IPv6 TCP and UDP packets, and IPv4
header
VLAN support: up to 64
Windows NIC teaming: up to eight teams per system, eight ports per team
Jumbo frame support: up to 9600 bytes
Header data split (HDS) feature for advanced link layer
Receive side scaling (RSS) feature for advanced link layer
TCP segmentation offload (TSO) and large send offload (LSO)
Link aggregation (NIC teaming)
Priority-based Flow Control (IEEE 802.1Qbb)
Enhanced Transmission Selection (IEEE 802.1Qaz)
Data Center Bridging eXchange Protocol (DCBX)


Managing CNAs
The Brocade CNAs for IBM System x can be managed by using either the Brocade
Command-Line Utility (BCU) that is installed with the drivers on the system, or the GUI-based
Host Connectivity Manager (HCM). HCM utilizes a client-agent based model. The HCM agent
is also installed with the adapter drivers. The HCM GUI client can be installed either on the
same system that the CNAs are installed on or on a separate management system.

Figure 7 shows HCM managing a Brocade dual-port CNA.

Figure 7 Brocade Host Connectivity Manager

Context-sensitive menus are used to configure devices within HCM.

Brocade CNAs support NIC teaming, which virtualizes two or more physical network
interfaces into a single logical interface. In essence, the driver advertises the same MAC and
IP address of all teamed interfaces. Although the following two types of teaming methods are
available, this configuration uses active-passive:
Active-Active
This method transmits across all devices in the team in a round-robin method; it provides
redundancy and load balancing. This method can be used only if the teamed interfaces
have uplinks to the same physical switch (or multiple switches) that support Brocade
multi-chassis trunking (MCT)
Active-Passive
One physical interface in a teamed group, called the primary port, is active while the other
interfaces are on standby for redundancy. If the primary port stops, a secondary port is
chosen to be the next primary.


The active-passive configuration can be in either of the following modes:
– Failback mode: If the original port designated to be the primary goes down and comes
back up, it becomes the primary port again
– Failover mode: The secondary port that became the primary stays as the primary port
until it fails.

The current Fast Track configuration supports only active-passive mode. Typically, all settings
such as MTU size, link speed, port VLAN ID, and others on the interfaces to be teamed must
match. Up to eight separate teams can be created on a single system with up to eight ports
per team. All ports must be from Brocade CNAs.

Note: FCoE traffic is not supported when active-active NIC teaming or bonding is used.

To configure NIC teaming and VLANs over a NIC team, right-click on the server level,
localhost in the example, and select Teaming, as shown in Figure 8.

Figure 8 Creating a team

Use the Teaming Configuration window to create, destroy, or configure NIC teams.

VLANs for these teams can also be added and deleted. In the next example, a team named
Test is created with two ports in failover mode. A pass-thru VLAN 0 always exists; this is the
device interface that forwards untagged packets. In addition, a device interface for VLAN
ID 10 has also been created for this NIC team. Packets sent out of this device are IEEE
802.1q, tagged for VLAN 10.


See Figure 9.

Figure 9 Configuring the teaming

In Windows, open Network Connections. Figure 10 shows that two new device interfaces are
listed: Local Area Connection 7, which is TEAM#Test and represents the pass-thru VLAN 0;
and Local Area Connection 8, which represents VLAN ID 10 for the NIC team we just created.

Figure 10 Network connections


Windows Device Manager
Jumbo frames and other options such as IPv4 Checksum Offload, TSO, LSO, RSS, HDS,
VLAN support, and VLAN ID can be configured as follows, before building teams of the
devices:
1. Open the Windows Device Manager, right-clicking the Brocade adapter you want to
configure.
2. Select Properties.
3. Click the Advanced tab.

Although VLAN ID can be configured in the Advanced property tab; it is not recommended in
the context of configuring your Microsoft Fast Track environment. Multi-VLAN configuration
should be done using either BCU or HCM and not Windows Device Manager. See Figure 11.

Figure 11 Device Manager

Related links
These documents provide additional information regarding setup and configuration.
Brocade 10Gb CNA for IBM System x at-a-glance guide
http://www.redbooks.ibm.com/abstracts/tips0718.html?Open
Brocade 2-port 10GbE CNA drivers and documentation for IBM
http://www.brocade.com/services-support/drivers-downloads/CNA/IBM.page
IBM ServerProven® compatibility
http://www.ibm.com/systems/info/x86servers/serverproven/compat/us/xseries/lan/m
atrix.html
Brocade 2-port 10GbE CNA for IBM System x product page
http://www.ibm.com/systems/storage/product/systemx/cna/index.html


Appendix C: IBM Ethernet Switch B24X
The IBM Ethernet Switch B24X is a compact high-performance, high-availability, and
high-density 1 RU switch specifically designed for mission-critical data center requirements.
This switch provides 24 10/1 GbE (SFP+) ports plus four 10/100/1000 MbE (RJ45) ports of
connectivity in an ultra-low-latency, cut-through, non-blocking architecture. The IBM Fast
Track configuration uses the B24X for storage connectivity for a high-speed iSCSI storage
network, eliminating the need for multiple aggregated 1 GbE networks that would normally be
required.

Note: Make sure you are running firmware R04.2.00a or higher to support the
configurations required for Fast Track.

Features are as follows:
Capable of full Layer2 and IPv4 Layer3
24x ports of dual-speed 10/1 GbE SFP+ ports plus 4x 10/100/1000 MbE RJ45 ports
Capable of line speed, full-duplex throughput on all ports—488 Gbps total switch capacity
and ultra-low latency (1.5 micro-seconds)
1+1 redundant, hot-swappable and load-sharing 300 W ac power supplies for high
availability
Hot-swappable, resilient triple fan assembly with automatic speed control for efficient
front-to-back cooling: 2+1 redundant fans in assembly
Industry standard command-line interface (CLI)

For a full list of features, see the following resources:
IBM Ethernet Switch B24X IBM System x at-a-glance guide
IBM Ethernet Switch B24X product page
http://www.ibm.com/systems/x/options/networking/ethernetb24x/index.html

Managing B24X
The IBM Ethernet Switch B24X uses an industry-standard CLI for management. This CLI can
be accessed over Telnet or SSHv2. Fault management and monitoring can also be
accomplished by using IBM Systems Director software. In addition, the switch supports
industry-standard SNMP.

For additional details about configuring the switch, see “Related links” on page 34. Also, see
the following resources:
IBM Ethernet Switch B24X Installation and User's Guide
Release Notes that are included with each firmware update for the switch


Fast Track configuration
Example 1 is a sample configuration file for an IBM Ethernet Switch B24X that is configured
as outlined in this document. This configuration file is for one of two B24Xs that are deployed
in a redundant fashion; the second B24X can be configured similarly.

Example 1 Fast Track configuration
Current configuration:
!
ver 04.2.00
!
!
global-stp
!
!
vlan 1 name DEFAULT-VLAN by port
spanning-tree
!
vlan 10 name iSCSI by port
tagged ethe 13 to 24
untagged ethe 1 to 12
spanning-tree
!
vlan 30 name Prod_Cluster_Private by port
tagged ethe 13 to 14 ethe 17 to 24
spanning-tree
!
vlan 40 name Prod_Cluster_Live_Migration by port
spanning-tree
!
Vlan 50 name Prod_VM_Network by port
router-interface ve 50
spanning-tree 802-1w
spanning-tree 802-1w priority 0
spanning-tree 802-1w ethe 14 admin-pt2pt-mac
!
vlan 60 name Management_Network by port
untagged ethe 25 to 26
router-interface ve 60
spanning-tree 802-1w
spanning-tree 802-1w priority 0
spanning-tree 802-1w ethe 14 admin-pt2pt-mac
!
vlan 70 name Management_Cluster_Private by port
spanning-tree
!
vlan 80 name SQL_Cluster_Private by port
spanning-tree
!
!


!
!
jumbo
hostname IBM_B24X_Bot
router vrrp-extended
snmp-server community ..... rw
interface ethernet 13
spanning-tree 802-1w admin-pt2pt-mac
link-aggregate configure key 10001
link-aggregate active
!
link-aggregate configure key 10001
link-aggregate active
!
spanning-tree 802-1w admin-edge-port
!
!
!
!
!
!
!
!
!
!
!
spanning-tree 802-1w admin-pt2pt-mac
!
!


!
interface ve 50
ip address 192.168.50.3 255.255.255.0
ip vrrp-extended vrid 50
backup
advertise backup
ip-address 192.168.50.1
enable
!
interface ve 60
ip address 192.168.60.3 255.255.255.0
ip vrrp-extended vrid 60
backup
advertise backup
ip-address 192.168.60.1
enable
!
!!
!
end

Related links
These documents provide additional information regarding setup and configuration.
IBM Ethernet Switch B24X at-a-glance guide
IBM b-type Data Center Networking: Product Introduction and Initial Setup
http://www.redbooks.ibm.com/abstracts/sg247785.html?Open
IBM b-type Data Center Networking: Design and Best Practices
http://www.redbooks.ibm.com/abstracts/sg247786.html?Open
IBM System Networking support page for IBM Ethernet Switch B24X (firmware and
documentation)
http://www.ibm.com/support/entry/portal/Overview/Hardware/Systems/Systems_netwo
rking/Ethernet_Switches/IBM_Ethernet_Switch_B24X_(4002-X2A)
IBM Ethernet Switch B24X product page
http://www.ibm.com/systems/x/options/networking/ethernetb24x/index.html


Appendix D: Worksheets
Use the worksheets in this section to help you plan the network configuration (Table 3 through
Table 10 on page 38).

Table 3 VLAN 10
VLAN 10 addresses (iSCSI storage) IP addresses

N6210 Controller-A

N6210 Controller-B

Management Server 1 - NIC1

Management Server 2 - NIC1

SQL Server VM1 (Coming off FT team VLAN10)


Hyper-V Production Server 1 - NIC1








Production VMs (if needed from FT team VLAN10)


Table 4 VLAN 20
VLAN 20 addresses (iSCSI storage) IP addresses

N6210 Controller-A

N6210 Controller-B

Management Server 1 – NIC2

Management Server 2 – NIC2



Hyper-V Production Server 1 – NIC2








Production VMs (if needed from FT team VLAN20)

Table 5 VLAN 30
VLAN 30 addresses (Production Cluster Private) IP addresses

Hyper-V Production Server 1









Table 6 VLAN 40
VLAN 40 addresses (Production Live Migration) IP addresses









Table 7 VLAN 50
VLAN 50 addresses (VM Production) IP address range

Routing IP Address (Virtual Routing Address)

Routing IP Address (Top Switch)

Routing IP Address (Bottom Switch)


Table 8 VLAN 60
VLAN 60 addresses (Management) IP addresses

Routing IP Address (Virtual Routing Address)

Routing IP Address (Top Switch)

Routing IP Address (Bottom Switch)

N6210 Controller-A

N6210 Controller-B

Management Server 1

Management Server 2

Management Cluster

SQL Server VM1

SQL Server VM2

SQL Server Cluster

SQL Server Cluster Service

SQL MSDTC Service (Optional)

System Center Operations Manager

System Center Virtual Machine Manager

System Center Self Service Portal 2.0









Hyper-V Production Cluster

Table 9 VLAN 70
VLAN 70 addresses (Management Cluster Private) IP addresses

Management Server 1

Management Server 1

Table 10 VLAN 80
VLAN 80 addresses (SQL Cluster Private) IP addresses

SQL Server VM1

SQL Server VM2


Appendix E: SnapVault sample code
Example 2 shows the code to initiate SnapVault transfer between volumes.

Example 2 Initiate SnapVault transfer between volumes
#Sample Code to initiate SnapVault transfer between volumes
#This can be called from within SnapManager for Hyper-V as a method to backup up
Snapshots off of the #production volumes.
#
Import-Module DataONTAP
$contr_N6210a=Connect-NaController -Name N6210a -RPC
$contr_N6210b=Connect-NaController -Name N6210b -RPC
$snaplist1=Get-NaSnapShot -Name vProd2OS -Controller $contr_N6210a
Start-NaSnapvaultSecTransfer -SecondaryPath /vol/vProd2OS_SV/qProd2OS_SV
-PrimarySnapshot $snaplist1[0].Name -Controller $contr_N6210b


The team who wrote this paper
Scott Smith is an IBM System x Systems Engineer working at the IBM
Center for Microsoft Technology. Over the past 15 years, Scott has worked
to optimize the performance of IBM x86-based servers running the
Microsoft Windows Server operating system and Microsoft application
software. He has extensive experience in helping IBM customers
understand the issues that they are facing, and developing solutions that
address them.

Thanks to the following people for their contributions to this project:
Steven Tong, Brocade
Bryan Hoke, Brocade
David Ye, IBM
Daniel Ghidali, IBM
Marco Rengan, IBM
David Watts, IBM Redbooks®
Linda Robinson, IBM Redbooks
Diane Sherman, IBM Redbooks

Now you can become a published author, too!
Here's an opportunity to spotlight your skills, grow your career, and become a published
author—all at the same time! Join an ITSO residency project and help write a book in your
area of expertise, while honing your experience using leading-edge technologies. Your efforts
will help to increase product acceptance and customer satisfaction, as you expand your
network of technical contacts and relationships. Residencies run from two to six weeks in
length, and you can participate either in person or as a remote resident working from your
home base.

Find out more about the residency program, browse the residency index, and apply online at:
ibm.com/redbooks/residencies.html

Stay connected to IBM Redbooks
Find us on Facebook:
http://www.facebook.com/IBMRedbooks
Follow us on Twitter:
http://twitter.com/ibmredbooks
Look for us on LinkedIn:
http://www.linkedin.com/groups?home=&gid=2130806
Explore new Redbooks publications, residencies, and workshops with the IBM Redbooks
weekly newsletter:
https://www.redbooks.ibm.com/Redbooks.nsf/subscribe?OpenForm
Stay current on recent Redbooks publications with RSS Feeds:
http://www.redbooks.ibm.com/rss.html


Notices

This information was developed for products and services offered in the U.S.A.

IBM may not offer the products, services, or features discussed in this document in other countries. Consult
your local IBM representative for information on the products and services currently available in your area. Any
reference to an IBM product, program, or service is not intended to state or imply that only that IBM product,
program, or service may be used. Any functionally equivalent product, program, or service that does not
infringe any IBM intellectual property right may be used instead. However, it is the user's responsibility to
evaluate and verify the operation of any non-IBM product, program, or service.

IBM may have patents or pending patent applications covering subject matter described in this document. The
furnishing of this document does not give you any license to these patents. You can send license inquiries, in
writing, to:
IBM Director of Licensing, IBM Corporation, North Castle Drive, Armonk, NY 10504-1785 U.S.A.

The following paragraph does not apply to the United Kingdom or any other country where such
provisions are inconsistent with local law: INTERNATIONAL BUSINESS MACHINES CORPORATION
PROVIDES THIS PUBLICATION "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESS OR
IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF NON-INFRINGEMENT,
MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Some states do not allow disclaimer of
express or implied warranties in certain transactions, therefore, this statement may not apply to you.

This information could include technical inaccuracies or typographical errors. Changes are periodically made
to the information herein; these changes will be incorporated in new editions of the publication. IBM may make
improvements and/or changes in the product(s) and/or the program(s) described in this publication at any time
without notice.

Any references in this information to non-IBM websites are provided for convenience only and do not in any
manner serve as an endorsement of those websites. The materials at those websites are not part of the
materials for this IBM product and use of those websites is at your own risk.

IBM may use or distribute any of the information you supply in any way it believes appropriate without incurring
any obligation to you.

Information concerning non-IBM products was obtained from the suppliers of those products, their published
announcements or other publicly available sources. IBM has not tested those products and cannot confirm the
accuracy of performance, compatibility or any other claims related to non-IBM products. Questions on the
capabilities of non-IBM products should be addressed to the suppliers of those products.

This information contains examples of data and reports used in daily business operations. To illustrate them
as completely as possible, the examples include the names of individuals, companies, brands, and products.
All of these names are fictitious and any similarity to the names and addresses used by an actual business
enterprise is entirely coincidental.

COPYRIGHT LICENSE:

This information contains sample application programs in source language, which illustrate programming
techniques on various operating platforms. You may copy, modify, and distribute these sample programs in
any form without payment to IBM, for the purposes of developing, using, marketing or distributing application
programs conforming to the application programming interface for the operating platform for which the sample
programs are written. These examples have not been thoroughly tested under all conditions. IBM, therefore,
cannot guarantee or imply reliability, serviceability, or function of these programs.

© Copyright International Business Machines Corporation 2011. All rights reserved.
Note to U.S. Government Users Restricted Rights -- Use, duplication or disclosure restricted by
GSA ADP Schedule Contract with IBM Corp. 41

This document REDP-4771-00 was created or updated on July 22, 2011.

Send us your comments in one of the following ways: ®
Use the online Contact us review Redbooks form found at:
ibm.com/redbooks
Send your comments in an email to:
redbooks@us.ibm.com
Mail your comments to:
IBM Corporation, International Technical Support Organization
Dept. HYTD Mail Station P099
2455 South Road
Poughkeepsie, NY 12601-5400 U.S.A. Redpaper ™
Trademarks
IBM, the IBM logo, and ibm.com are trademarks or registered trademarks of International Business Machines
Corporation in the United States, other countries, or both. These and other IBM trademarked terms are
marked on their first occurrence in this information with the appropriate symbol (® or ™), indicating US
registered or common law trademarks owned by IBM at the time this information was published. Such
trademarks may also be registered or common law trademarks in other countries. A current list of IBM
trademarks is available on the Web at http://www.ibm.com/legal/copytrade.shtml

The following terms are trademarks of the International Business Machines Corporation in the United States,
other countries, or both:

IBM® Redbooks (logo) ® System Storage®
Redbooks® ServerProven® System x®

The following terms are trademarks of other companies:

Microsoft, Windows, and the Windows logo are trademarks of Microsoft Corporation in the United States,
other countries, or both.

SnapVault, SnapRestore, SnapManager, SnapDrive, Data ONTAP, NetApp, and the NetApp logo are
trademarks or registered trademarks of NetApp, Inc. in the U.S. and other countries.

Intel, Intel logo, Intel Inside, Intel Inside logo, Intel Centrino, Intel Centrino logo, Celeron, Intel Xeon, Intel
SpeedStep, Itanium, and Pentium are trademarks or registered trademarks of Intel Corporation or its
subsidiaries in the United States and other countries.

Other company, product, or service names may be trademarks or service marks of others.


IBM System x Private Cloud Offering, Advanced Configuration: Architecture and Implementation Guide

More Related Content

What's hot (19)

Viewers also liked (6)

Similar to IBM System x Private Cloud Offering, Advanced Configuration: Architecture and Implementation Guide (20)

More from IBM India Smarter Computing (20)

Recently uploaded (20)

IBM System x Private Cloud Offering, Advanced Configuration: Architecture and Implementation Guide