Saving IT Costs with ‘Base’ Systems Management Part 1 – Deployment
Systems Management Part 1 – Deployment
I have read many articles, posts and sales pitches, and there are many opinions on what an IT department should do in relation to their computing infrastructure. When it comes to ROI (Return on Investment), there is almost no current figures for systems management that are current, so it is very hard for companies to justify the purchase of systems management software as well as the installation and maintenance costs.
In this series, I will try and show what the best solution to start with when it comes to managing the computers and servers in companies with 350+ client nodes.
In the years I have been an IT professional, I have put all of my focus towards systems management and streamlining the IT infrastructure. I am always trying to find the most efficient, cost effective way to do things. I guess you could say I don’t like to work hard….just smart. In most cases I end up working myself out of work because of the efficiencies I put into place for all of my clients. Not the best business model but it is what I enjoy doing and the clients are always happy.
So let’s get into the information I want to present for this topic.
Deployment of new computers
I am going to give some general figures for this piece. Lets say you buy a new computer from your preferred vendor. 90% of the time you will want to put on your own operating system and deploy it to your workers. This can take some time. let’s say to get a computer up and running from bare metal with all the relevant security patches for Windows XP or Vista takes up to 4 hours. In some cases it will be much more but we will just use 4. Multiply this by the amount of computers you are deploying and you get some pretty staggering costs for tech time.
Problems with this method
- Without great documentation on how to build these computers, then many will end up being built slightly different and can actually raise the support costs due to the complexity and difference of the environment.
- 100 computers is 400 hours of tech time. The math is not very good here.
Common operating system images per model
Let’s say you have a common image per hardware platform, and you have 7-10 different computer models. This will bring down the time per deployment but managing the amount of images can be a daunting task. I would suggest that the time savings for this scenario could be up to 40%.
This will ensure that all computers have the same image per model which helps in lowering day to day support costs
Hardware Independent Imaging
This term means that there is one image that will span across all of the hardware models for the operating system being deployed. This is the most efficient method for deploying all computers in a company’s infrastructure. This method requires a good deployment tool and the person using it to have training and skills to be able to use it. In my opinion this will likely save up to 80% of deployment time
Benefits
- Reduced deployment time up to 85% depending on amount of computers
- All computers should have the same image. Support costs become much lower
- If a computer has a problem then all of them should. If not, something is different about the problem asset and should just be rebuilt. Distinguishing the problem should be very easy as well.
- All computers should have the same patch level when it comes to security
In summary, there are a few methods of deploying computers throughout your company, but there are some factors that need to be evaluated if costs are a concern. Obviously the best method is to streamline and automate the deployment method as much as possible. With Windows 7 coming later this year, it is all the more necessary to have a deployment tool well established in your company and technical staff that are trained and capable of using the tools to be ready for the new operating system.
Here is a link to the toolset I specialize in
For more information or if you have any questions, please feel free to email me at billasullivan@gmail.com
