This is my first post in a series that I am intending on writing regarding virtualisation projects aimed at small and medium businesses but a lot of the information may be relevant to larger companies as well. I am intending to cover all aspects from the start of the project, implementation, completion and monitoring.
What I would like to achieve by the end of this series is to give an IT manager / Technical Project Leader some information to help them with decisions, to help them work intelligently with or without a partner and give them food for thought with regards to elements they should be considering throughout the whole project cycle.
This series of blog posts is not aimed at your virtualisation professionals and isn’t aimed to be a in-depth technical article or how to guide. Saying that if there are any virtualisation professionals, IT Manager or Technical project manager reading this I would be happy for your input on the subject. DM me on twitter @virtualisedreal or email barry (at) virtualisedreality.com if you would like to assist with this series of blog posts.
My background, I have worked for a VMware Enterprise partner for a number of years and prior to this I was sitting on the other side of the desk working in the IT department of a medium size business.
I am going to start out with a few assumptions, I am going to be assuming that you have already decided on your virtualisation technology being VMware, although a lot of this is also transferable to Xen and Hyper-V I would imagine. I am going to assume the IT department is a small department with limited resources and no in house virtualisation specialists. I am going to assume you know the fundamentals of virtualisation.
This series of blog posts will consist of the following 6 subjects. I am looking for any input throughout the series that you think maybe of use to others reading/
- Post Project Planning
- Finding a Virtualisation Partner
- Solution Design
- Solution Implementation
- Post Project
- On-Going Monitoring and Maintenance
Pre Project Planning
You have read all the press information and all the sales material and have decided that virtualisation is for you. You know the benefits and have decided that VMware is the product for you. You may have even completed some kind of toe dipping exercise with ESXi or VMware server. You are now contemplating starting a roll out and you are starting to plan your solution and more than likely working with a partner on some or all of your proposed project.
Know what you want to achieve
This sounds very obvious but knowing what it is that you want to achieve is something that is often overlooked, the project goal ends up being we want to be virtualised and not what you started out to achieve. The technology should help you reach your goal and not be your goal.
A good place to start would be a mind map, white board session or list of what you want to achieve at the end of the project. This means you can ensure these goals are met and that you are heading in the right direction throughout the project. This will also help you dictate your requirement to your partner.
Some common reasons for virtualisation are listed below, but by no means is this an extensive list, a lot of decisions will be business driven and not included with these common goals.
Hardware Refresh / New servers required
Your existing hardware is tired or about to become out of warranty or you have a new server requirement and you want to use this opportunity to move to a virtualised environment.
Reduce Power Consumption
You are looking to reduce the overall power consumption within your datacenter.
Better Performance / Reliability
Users are starting to complain that applications or services are running slowly or unreliably and you need to improve the performance / Reliability of applications / services.
Lack of Space
You are rapidly growing and are fast running out of space for more physical servers
Your current DR strategy is lacking or non existent.
You backups regularly fail and you don’t have the time / resource to be able to test your backups regularly.
You have no way of centrally monitoring your servers and need to improve the way you monitor performance and detect issues.
Know your current enviroment
This sounds even more obvious than the previous subject, but you would be amazed by the amount of people that don’t actually know what they currently have or what each server does.
Your partner will be able to assist with this during a capacity planning exercise but before you even get to this stage it would be worth ensuring you have a good understanding of your environment.
It would be worth starting a spreadsheet or similar document with the following information and anything else you think maybe useful.
- Server Name
- IP Address
- Make and Model
- Full Specification
- Remaining Warranty
- Storage Utilisation
- Available downtime (for the Virtualisation Project)
- Responsible person
This document will not only assist during the planning stages but will help you or a partner identify hardware that maybe re-utilised during the virtualisation project. It is also worth noting any Switches, SANS and UPS’s and Backup devices for this purpose as well. Alongside this information it would be worth noting performance characteristics such as “This server is currently struggling to meet it’s requirements” or “This server was recently purchased and is over specified”
Also networking diagrams including speed and type of links between sites, to be used for DR and network layout including VLANs will be key in the design of your virtualised infrastructure.
Know your budget
It is important to know a rough idea of what budget you are working to. At the end of the day the requirement will in turn lead the needed solution but without any idea of budget it will be very difficult going forward to decide on the best solution for you.
I will leave this post here for now as I am aware I have gone on a lot already, but keep checking back for the next update next week and feel free to ask any questions or send me an email if you can offer any assistance with this subject.