Gavin Payne is a principal architect for Coeo, a SQL Server professional services company, and a Microsoft Certified Architect and Microsoft Certified Master. His role is to guide and lead organisations through data platform transformation and cloud adoption programmes.
In this article, we’ll see how you can use a free tool from Microsoft to identify the servers in your environment prior to starting data-centre wide IT projects.
Change is the new norm
In the last few years, not only has the pace of change in IT increased but also the scale of change. To borrow a cliché: whether it’s migrating physical servers to virtual servers, on-premises services to the cloud, or upgrading software – the only constant is the fact that everything keeps changing.
When we think about the types of change mentioned above its possible that they’re going to affect a large number of systems in an organisation. Replacing Windows Server 2003 for some environments means refreshing the majority of its servers, the same might happen when you consolidate your physical workloads to a virtual platform.
If that’s a challenge you’re facing, then the first step in any multi-server upgrade project should be defining your scope.
The purpose of a scope
A scope tells you what your project needs to consider and what it can ignore. In the context of a server environment, this might tell us which servers you have to upgrade and which you can leave as is. Essentially, it defines the size of the problem.
How to define a scope
If you’ve got a small enough environment, you might be able to identify all of the in-scope servers for a project from information you already know. If not, you’re going to need to perform a discovery and have something tell you what’s in your environment. Some might be surprised that information like that doesn’t already exist, but the reality is if it does then it’s typically not trusted to be up to date so still needs collecting again.
Discovering your environment with the Microsoft Assessment and Planning Toolkit
Microsoft’s free Assessment and Planning Toolkit, known as MAPS, is a simple but powerful tool that lets you discover all of the servers, their configuration and their workload amongst many other areas of detail.
It’s available to download from its TechNet home page here:
I use it to do the following:
- Discover the servers in a Windows domain or set of IP address ranges
- Catalogue each server’s software and hardware configuration (versions, service packs etc)
- Where it can, delve deep into the server’s virtualisation or SQL Server configuration
- Collect workload profiles over a 7-20 day period using its internal performance monitor collectors
- Store all of the above in a single repository with a graphical interface
- Export the summarised data it collected above to a series of Excel files for me to analyse
It can do far more, but for a mid-size organisation unneeded with exactly what they have in their IT environment, how its configured and how hard it works this can often be the first time they see this kind of information in a single report.
Defining your scope
One of the first outputs from the MAPS tool is the name of every server in the environment it scanned – that’s version 1 of your scope.
Next, you need to start thinking about servers you’re planning to retire before the project finishes. Then, you should consider any servers in your data centre that you don’t own or manage – and so on. Your objective is to make your scope as small as you legimately can. No one wants to do un-necessary work but at the same time you can’t ignore servers for no reason.
Hopefully this article has either introduced you to the free Microsoft MAPS tool, or if you’re already familiar with it then hopefully you’ve learnt something from me explaining one of the ways I use it.