How to virtualise SQL Server
Ways to prepare for a virtualised environment
By Chike Okoye | Techworld | Published: 11:10, 09 March 2009
With VMware becoming more common place in the work place, many organisations are still reluctant to deploy production SQL servers on VMware infrastructure. This has been understandable in the past but VMware environments are now more robust and should be able to handle SQL Server.
It's worth taking the plunge; there are some very good reasons to move some of your SQL servers onto a virtual infrastructure.
Probably the biggest selling point of virtualisation is the ability it provides to consolidate mission-critical applications and infrastructure services onto fewer highly scalable, highly reliable enterprise-class servers with organisations able to achieve up to 60-80 percent server utilisation thus reducing TCO across their IT infrastructure.
Related Articles on Techworld
Simplified system management
VMware allows organisations to provision new applications and respond to system change requests in minutes not days and conduct zero-downtime hardware maintenance without waiting for maintenance windows. Its dynamic load balancing capability allows for improved flexibility and responsiveness.
Higher Availability and improved service levels.
Being able to precisely control system resources granted to virtual machines and run them across multiple processors with Virtual SMP (optional) means that resource intensive workloads like SQL server can be run on multiprocessor virtual machines. Low-cost virtual machine clusters providing hardware and software fault tolerance can be deployed on physical clustered systems protecting critical data.
Lower Cost of Disaster Recovery Capability
Using VMware organisations can create a unified disaster recovery platform using ESX Server virtual machines as standby servers. A single x86 system can host multiple disaster recovery virtual machines maintained in a hot or cold state. Stream lining disaster recovery management, increasing availability, reducing recovery time and lowering hardware and operational costs by reducing the need for one-to-one mapping of production and disaster recovery servers.