Quick Blog Description: In this series of postings I will be detailing how to use System Center Virtual Machine Manager to deploy complex applications / services. I will be demonstrating this capability by deploying a System Center Orchestrator instances consisting of multiple virtual machines hosting the Orchestrator components. While my example focuses on Orchestrator, the concepts and most of the core configuration will be applicable for just about any complex service or application deployment with SCVMM.

Introduction to Service Templates: Essentially, Service Templates allow us to rapidly deploy a pre-configured services (application), consisting of one or more virtual machines, managed as a single entity, which can be scaled to match demand. A simple example may be a two tiered application consisting of a SQL Server, and a load balanced web server. As demand for the application increases, the web interface may be scaled out into a three or four NLB cluster. As demand decreases, the service can be scaled back to the initial two node cluster.

We configure Service Templates through the use of many resources already familiar to the SCVMM administrator such as virtual hard drives, VM Templates, and other VMM library resources. The service template allows us to configure how many VM’s are initially deployed as part of the service deployment, configure applications including SQL Server to be installed at deployment time, as well as to configure other scripts to be execute throughout the deployment. Finally service templates allow us to define thresholds or minimum and maximum instance sizes at which the service can be scaled.

My goal with this blog series is to not only provide an example service template, but to also deep dive on the items that need to be completed in order to create your own service templates / deployments. Rather than have one large post to complete this task, I am going to break this post up into the following chunks.

I hope that you enjoy this series and that you can take some of the lessons that I have learned about rapid service deployment and apply them for your own benefit.

Scenario or Example for this Blog:

I will be deploying a multi-tier Orchestrator Instance that can be scaled as necessary. My initial deployment needs to consist of one server which contains all Orchestrator roles (Management Server, Database Server, Web Components, Runbook Server). I will then deploy an additional four servers which will each contain an additional Runbook server.

So my initial deployment will look thusly:

  • Server 1: All Orchestrator Components (Database Server, Management Server, Web Components, and Runbook Server)
  • Server 2: Runbook Server 2
  • Server 3: Runbook Server 3
  • Server 4: Runbook Server 4
  • Server 5: Runbook Server 5

In order to achieve this goal, the following actions need to occur at initial deployment:

  • 5 VM’s are deployed
  • SQL Server is deployed to VM1
  • All Orchestrator components are deployed VM1
  • Four additional Runbook servers are deployed to the remaining VM’s (VM2 – VM5)

To close out this introduction I am going to include a video demonstration of this service deployment in action. If you like what you see, stay tuned for the next few blog posting in which I will detail how this deployment was configured.

Direct Link to Video - http://youtu.be/_QQLPvURntg