How Windows Server 2012 improves DC sustainability

Hardware requirements for Windows Server 2012

Windows Server 2012 was officially released online on September 4, allowing customers to finally buy the latest server operating system. However, since the official version has just been released, many users are unfamiliar with the latest version of Windows Server and don't know the features of the new version.

In fact, the latest version of Windows Server 2012 and other Microsoft products offer a wide range of high-availability options, covering everything from the infrastructure to the application. Here is a brief introduction to the different technologies and guidelines that you can use to avoid every possible failure point and provide you with an ongoing, available data center to prevent planned and unplanned outages.

For hardware requirements:

Before we deploy to servers and services, we first need to determine that the data center and physical infrastructure can maintain availability after some components fail, or they must be taken offline and maintained. The data center itself must have a backup power supply to prevent sudden blackouts and other conditions. We can use generators or batteries, and each server must have a redundant power supply and connect to different electrical outlets.

Network facilities must also be fully redundant, including switches, routers, and hardware balanced load devices that we often use. The NIC should be bundled and should have redundant paths to all networks, including connections to the Internet.

Redundant array of independent disks (RAID) technology should be used for storage

Prevent disk failure and data should be copied or mirrored to the secondary array. Do a backup of your data.

In addition, you can deploy Multipath I/O (MPIO) to provide multiple communication paths to storage devices.

If Internetsmall Computer Systems Interface (iSCSI) storage is used, iSCSI target itself should create clusters to reduce downtime. Even if every component of the data center is highly available, it is still necessary to realize that natural disasters can damage the entire site, so you need to consider building secondary data centers through multi-site clustering or replication technology to achieve disaster recovery.

Windows Server 2012 Infrastructure

Requirements for server infrastructure:

Once the data center is ready, we also need to ensure that all critical server infrastructure components are highly available.

First, make sure that each server role has multiple instances that can provide redundancy for all services.

In Active Directory, different high availability options are available for different roles. Active Directory supports backup and restore, multi-site load balancing, and recovery of deleted objects through the Active Directory Recycle Bin feature.

In addition, we can deploy read-only domain controllers in less secure locations or branch offices.

Active directorycertificate Services (AD CS) can support failover clustering. Active Directory Federation Services (AD FS) can support Cross-site replication and create SQL mirrors for your own databases. Active Directory Lightweight Directory Services (AD LDS) can also support cross-site replication and backup and restore. The database for Active Directory Rights Management Services (AD RMS) server can take advantage of the high availability of SQL (failover clustering or log shipping is available). The license server can use the Network Load Balancing domain Name System (DNS) to use the round robin algorithm to send clients to different DNS servers, which provides simple load balancing through redundant servers.

NLB is a software-based solution that delivers high availability and scalability by dispersing traffic across multiple redundant servers. This technique is typically used for Web sites hosted on Internet Information Services (IIS) for fully consistent data on each node, and for changes that are not very frequent in the server role.