Its that time of year when everyone starts to predict strategic IT technologies for 2011, so we thought we would have a punt too! Many organisations we have spoken to over the last calendar year are still planning on fattening out their virtualisation estate over the course of the next year to include, high availability, desktop and greater consolidation ratios, whilst being mindful of the ability to adopt cloud. And because of the close nature between virtualisation and cloud technologies it is clear that management is one of the top priorities. I think it is therefore prudent that you should consider some of the management scenarios in your planning for next year:
Capacity planning and optimisation. Most people that I talk to when I ask them how they scaled their virtualisation infrastructure look blankly at me and have over provisioned like they did in the physical world. Capacity planning and optimisation tools will calculate just how much public or private infrastructure you need to run your apps and how best to arrange the VMs for maximum efficiency.
Virtualised I/O. Virtualised I/O -- usually in the form of 10 Gb converged Ethernet. Aside from benefits like reducing cabling and switch ports, I think the most interesting aspect of virtualised I/O is the ability of a physical server's personality to be moved to any other server in the data centre. In addition to the underlying network technology, the thing that makes this possible is integrated management of the server and data centre fabric.
Public and private cloud management. Management tools for managing virtualisation platforms will integrate with cloud services. Initially it will be about monitoring or provisioning. Later on, your system management infrastructure will facilitate the movement between internal and external resource pools.
Streamlining ongoing management of servers. If you manage a server the same way you always have, then the ongoing costs of maintaining a virtualised server will be the same as they always have been. In addition to standardising images and other process-related things, you should be automating configuration management.
Real-time automation of workload management. As server hardware becomes more powerful we'll naturally want to increase our VM density to achieve a lower cost per VM. The result is that individual servers will be increasingly critical to application availability.