Planning a High Availability Hyper-V Configuration

iSCSI SAN Storage Software / Application Solutions / IP SAN for MS Hyper-V / Planning a High Availability Hyper-V Configuration

Planning a High Availability Hyper-V Configuration

One of the many advantages to a fully virtualized environment is the ability to quick deploy similar or identical virtually machines onto the same server, allowing each virtual machine to be configured and used for its own distinct purpose. In production environments and/or in instances where you are hosting a web-facing application that perhaps is the backbone of your business, downtime can be very expensive and could even cause enough issues to cost your customers or the business itself.



In such a scenario, it is important to plan for a high availability configuration that ensures that the overall system remains up and responsive even if a piece of the infrastructure fails. Examples of failures that often occur but that can be planned for include network interface failures, Internet network connection outages, hard drive errors or failures, web server crashes, or application bugs that cause an environment to become unstable or completely down.

In any of those scenarios, businesses can plan for the failure up front, assuming you have the time and money to do so. Truly large-scale mission-critical applications that involve heavy financial, equipment, or (worse yet) human losses typically plan for high availability options at every single known single point of failure. RAID configurations are done across multiple hard drives to plan for drive failures. Multiple network points of redundancy are fed into the infrastructure so that if one drops, communications can continue. Web software or database software layers can be configured for high availability through software or hardware load balancing on the application tier or active-active/active-passive database server configuration on the database tier.

Organizations that have based their infrastructure around Microsoft Hyper-V quickly grow accustomed to a virtualized environment but, at the same time, become increasingly at risk of a single minor hardware or software failure bringing down not one physical server but many virtual servers, as all may be running on a single machine. If this single machine fails, all virtual servers onboard will also fail which actually can be viewed as an increased risk to virtualized hosting options.



Several planning steps can be undertaken to mitigate such a risk in order to ensure your Hyper-V virtualized environment is highly available. The first piece of advice is to ensure that your Hyper-V environment is configured in the Windows Failover Cluster mode, which requires at least two physical servers, which can be expanded up to a total of 16 nodes. In this configuration, it is important that your machines be configured identically so that testing the failover from one machine to another is an apples-to-apples comparison.

Just as with database servers, Hyper-V clusters can be configured in both an active-passive or active-active setup. Active-passive means that multiple servers are available, but only half of those are live and in use at any moment. The other half can be though to be on stand-by; in the event of a failure they can be switched into use. An active-active configuration is one in which half the environment is live, but the other half is in constant communications and is in sync with the life environment. Should the live environment go down, the other set is instantly brought online and users would experience little to no interruption.



The primary difference between the two configurations is one of performance and total downtime in the event of a failure. The final critical configuration in a highly available Hyper-V cluster is that of storage. Hyper-V clusters are required to run on shared storage, which typically involves the use of a SAN compatible with Windows Server 2008 R2 failover or an iSCSI configuration. With the proper amount of planning and preparation, it is possible to build a highly reliable, highly available clustered environment using Hyper-V virtual machines.

Download Hyper-V and High Availability Shared Storage Guide:

---

Live demo		Free 30-day trial		How to buy
Request now		Download free trial		Request a quote