Clustering is an indispensible technique to achieve maximum fault tolerance. A cluster is a collection of computer systems responsible for running a shared set of applications and services. The systems in the cluster are interconnected through hardware and software in such a way that in case any component in the cluster stops functioning, the workload will "failover" to, or be assumed by, another system in the cluster without a disruption of service.

Failover clustering clearly has an advantage in increasing network availability in the case of a hardware breakdown, but it also allows non-disruptive system maintenance for environments that have no time window when key applications such as email or web commerce can be taken offline without hurting business. In the case of geographically dispersed environments, clustering provides insurance against widespread disasters such as earthquakes or hurricanes.

Enhanced-capability editions of Windows Server 2008 include significant improvements in failover clustering such as easier deployment, management, and migration; enhanced support for multi-site clusters; improved security architecture and quorum models; stronger authentication and encryption; and more efficient tools for backup and recovery.

Progent offers a full range of Windows Server 2008 Server consulting services and can assist businesses of any size to design, implement, and maintain cost-effective failover clusters that provide high availability, security, and performance and that add significantly to the business value of your information network.

New and Improved Features of Windows Server 2008 Failover Clustering
The failover clustering capabilities incorporated into Microsoft Windows 2008 Server represents more than a decade of Microsoft's realworld experience in developing clustering technology. The Enterprise and Datacenter Editions of Microsoft Windows Server 2008 offer a number of new clustering services and capabilities including:

More Convenient Management Interfaces
Windows Server 2008 includes Microsoft Management Console (MMC) 3.0, a new easy-to-use interface for controlling network resources. The new Failover Cluster Administration Console is task oriented rather than cluster resource oriented. To make a file share highly available, for example, the administrator no longer has to create a group, a disk resource, an IP address resource, Network Name, configure IsAlive/LookAlive, configure preferred services, and set dependencies. Instead, the wizard simply asks for the Network Name. You do not necessarily need to assign an IP address because Windows Server 2008 failover clustering now supports DHCP.

Windows Server 2008 also supports remote management via the Remote Server Administration Tools (RSAT). You can manage multiple clusters throughout the organization from a single MMC. Windows Server 2008 Failover Clusters are also fully scriptable with Windows Management Instrumentation (WMI).

Built-in Cluster Validate Tool
Windows Server 2008 clustering includes a cluster Validate Tool, an expanded and integrated version of the ClusPrep tool released for Windows Server 2003 server clustering. As a part of the cluster configuration process, Validate runs a set of tests to determine whether the selected servers, cluster networks, and storage meet the requirements for failover clustering. Validate performs an inventory that includes BIOS information, operating systems, adapters, drivers and software updates; verifies cluster topology for redundancy; checks IP configuration details; and makes sures that storage can correctly handle simulated cluster actions. The Validate Tool helps prevents configuration errors, which is one of the most common problems encountered in failover clusters.

Support for Windows Server 2008 Server Core
Windows Server 2008 offers a Server Core installation option that supports failover clustering and provides a bare-bones environment for running distinct server roles. By limiting the services that need to be maintained and updated, this minimalist installation reduces support and management expense and enhances security by shrinking the server's attack surface. Server Core supports a variety of important roles including DHCP server, file services, print services, DNS server, Active Directory Domain Services, streaming media services, and Windows Server 2008 virtualization.

Better Storage and Backup Support
Improvements in storage and backup offer increased stability and availability for failover clusters. Support for both Master Boot Record (MBR) and GUID Partition Table (GPT) disks allows the creation of disks greater than 2 terabytes. Built-in logic initiates self healing automatically as long as the system can find either the disk signature in the Master Boot Record or the SCSI inquiry page 0x83 data (VPD).

Support for Storage Area Networking (SAN) solutions is improved for storage hardware that supports SCSI-3 SPC3–compliant SCSI commands for persistent reservation/release. All multi-path solutions are Microsoft Multipath I/O (MPIO). The Device Specific Module (DSM) provided by the storage vendor is responsible for sending PR registration information down each path to the storage system. Closer integration with Volume Shadow Copy Service (VSS) makes the backup process easier. Failover clustering in Windows Server 2008 has its own VSS.

Greater Expandability
With Windows Server 2008, x64-based failover clusters support up to 16 nodes in a single cluster, as opposed to the maximum of 8 nodes in Windows Server 2003. Windows Server 2008 failover clusters also now support GPT disks, which use the GUID partition table (GPT) disk partitioning system. A GPT disk allows up to 128 primary partitions and a much larger volume size. In comparison, MBR disks can support up to four primary partitions and a volume size limit of up to 2 terabytes. For greater reliability, GPT disks offer replication and cyclical redundancy check (CRC) protection of the partition table. The increased number of nodes plus support for GPT disks dramatically extends the scalability of larger volumes in cluster environments.

A New Quorum Model
Windows Server 2008 introduces a new failover clustering quorum model that incorporates a combination of Microsoft's earlier shared disk and majority node set models. In Windows Server 2008 failover clustering there are now four ways to establish a quorum:

• No Majority – Disk Only (same as Windows Server 2003 shared disk quorum)
• Node Majority (same as the Windows Server 2003 majority node set)
• Node and Disk Majority
• Node and File Share Majority

The quorum mechanism in Windows Server 2008 no longer requires a shared storage resource for the cluster to function. Quorum now refers to a number of votes which must equate to a majority of nodes. All nodes and disk resources have a vote. This avoids failure points in the old model, which required the disk to be available at all times so that disk failure meant cluster failure.

An Improved Security Model
Windows Server 2008 failover clustering incorporates a new security model that enhances its security and dependability. Changes include:

• Elimination of the need for a domain user account for the Cluster Service Account (CSA)
• Better logging and event tracing
• Transition from unsecure datagram remote procedure call (RPC) communications to TCP-based RPC communications
• Default enabling Kerberos authentication on all cluster network name resources
• The ability to audit access to the cluster service via either the Failover Cluster Management snap-in or the cluster command-line interface
• Ability to secure Inter-Cluster communications.

New Networking Capabilities and More Flexible Dependencies
Windows Server 2008 failover clustering includes a new networking model. Major improvements include:

• Enhanced support for geographically distributed networks
• The ability to place cluster nodes on different networks
• The ability to use DHCP server to assign IP addresses to cluster interfaces
• Improvements in the cluster heartbeat mechanism
• New support for IPv6

Administrators no longer have to stretch virtual local area networks across the wide area network to support geographically dispersed servers on different subnets. Failover cluster nodes can now be located on different subnets. Individual cluster nodes can be placed on separate, routed networks, which allows IP Address and Network Name resources to come online when services and applications fail over to remote nodes.

Cluster IP address resources can now get their addressing from DHCP servers as well as from static entries. The cluster heartbeat mechanism has moved from a UDP broadcast health checking mechanism to a UDP unicast communication that it uses a Ping-like Request-Reply process. This allows better security and more reliable packet sequence numbering. Windows Server 2008 supports IPv6 IP Address resources and IPv4 IP Address resources either alone or in combination in a cluster. Clustering also supports 6-4 and Intra-site Automatic Tunneling Addressing Protocol (ISATAP).

Progent's service professionals can also help you migrate from Windows Server 2003 cluster nodes. This migration requires experience because these two types of nodes cannot share the same cluster, which eliminates a rolling migration.

Progent's CISM, CISA, GIAC, and CISSP certified security engineers can assess the protection offered by your cluster environment and can recommend and implement affordable improvements. Progent's security experts can also help you design, document, and test business continuity and disaster recovery plans.

Other Fault Tolerant Solutions Offered by Progent's Consultants
Progent's Microsoft and Cisco consulting experts can help your company implement an affordable, high-availability IT architecture that resolves key problems covering a wide array of infrastructure technologies and operations such as:

24x7 Internet Access Architecture
Progent's Cisco-certified network consultants can show you how to build an affordable, 24x7 Internet connection architecture that can achieve 24x7 Internet connectivity via a wide array of fault tolerant Internet access technologies including non-stop BGP, automatic fail-over, and redundant ISPs. Progent offers Cisco Certified Internetwork Expert professionals to show you how to apply the most current technology for fault tolerant Internet connections to implement a cost-effective, completely redundant Internet network with automatic failover and other features to deliver fault tolerant Internet access. For details, visit 24x7 Internet Access Architecture

Non-stop Load Balancing
Progent provides high availability load balancing support that addresses network load balancing, load balanced applications, LAN/WAN backbone routing, and content delivery technology including Cisco Content Engine. High availability load balancing technologies for which Progent offers consulting services include Microsoft Windows Server 2003 and 2008 Network Load Balancing Manager, Citrix Metaframe and Presentation Server, Cisco CSS, Cisco Content Distribution Manager and LocalDirector 400 series, and F5 Networks BIG-IP. To read more, see Non-stop Load Balancing.

High-Availability Internet Data Centers
Progent's Data Center consultants can help you select high availability data centers that match the particular needs of your business. Fault tolerant collocation sites are particularly attractive to small companies due to of their low cost and simplified logistics compared to taking a do-it-yourself strategy. Attempting to build an on-site data center that offers even a minimum level of fault tolerance is too costly for typical small businesses. Progent can show you how to get all the benefits of high availability data centers and 24x7 colo sites by offering a complete range of consulting and support services. To learn more, visit Fault Tolerant Internet Data Centers Solutions.

Non-stop Inter-office Connectivity
Progent’s high availability inter-site connectivity and fault tolerant network connection professionals can help small companies achieve non-stop Interoffice connectivity via fully redundant system architectures featuring transparent failover. Progent’s Cisco-certified Cisco Certified Internetwork Experts can plan and deploy high-availability Inter-office networks based on non-stop EIGRP network routers or non-stop OSPF deployment of Cisco routers. Progent can demonstrate how multi-path connections and automatic fail-over can provide economical non-stop interoffice connectivity. To learn more, see Non-stop Inter-office Connectivity.

Disaster Recovery Consulting and Business Continuity Consulting Services
Progent's disaster recovery support and business continuity professionals can show you how to design a disaster recovery plan to prepare for an IT network catastrophe. Progent can help you develop a complete disaster recovery strategy that includes periodic disaster recovery evaluations and drills. Progent's Microsoft and Cisco-authorized professionals can also show you how to build an affordable, high-availability network solution that addresses reliability issues involving a broad array of network technologies and processes. For more information, go to Disaster Recovery Consulting Services.

Microsoft DPM Backup and Recovery Consultants
Progent's expert System Center Data Protection Manager engineers can help your company to create a practical business continuity plan, implement an affordable high-availability IT architecture, configure a Microsoft DPM-powered recovery system, comprehensively test your system, educate your administrative personnel or local service providers how to maintain Microsoft DPM, or handle extensive IT support outsourcing including off-site housing of Microsoft Data Protection Manager servers in Progent's secure data center facility. To find out more, see Data Protection Manager (DPM) Consulting Services.

Why Companies Select Progent
Since Progent is a Microsoft Gold Partner and provides specialized application consulting for a broad variety of small business applications, Progent is able to assist clients to arrive at efficient information technology solutions that substantially enhance the productivity of your information system. As a Registered Partner with Cisco Systems, Progent can provide the expertise of certified consultants and specialists with expertise in architecting, implementing and supporting networking environments based on Cisco products. The depth of Progent's technical know-how and Progent's appreciation for the requirements of small and midsize businesses make Progent the ideal resource for creating and running a cost-effective, dependable network.

How You Can Get Help from Progent's Consultants for Non-stop System Support
To ask Progent about consulting help for non-stop IT design, phone 1-800-993-9400 or e-mail fault-tolerant-help@progent.com.