Appendix B: Performing Infrastructure Remediation

Article
03/05/2008

Infrastructure remediation is one of the two most important tasks this feature team must perform. Remediation is a process that relies on a given series of steps. This process is illustrated in Figure 10.

Figure 10. Infrastructure Remediation process

This process includes the following steps:

Begin by identifying whether existing distribution points are available.

This would be the case if the organization were already using a system management tool or if it had deployed servers in a specific computer system deployment or stand-alone installation support role.
If no distribution points are available, identify whether existing servers have the capacity to double as a distribution role.

Because this role is primarily a file server role, existing file servers could possibly perform this additional role.
If no servers or distribution points are available, consider the purchase of additional servers in support of the deployment.

These servers can be placed permanently into the network, or, if the role will not be permanent, the organization can choose to use portable servers that can move from location to location with the Deployment feature team as the deployment progresses.
After identifying the target servers, validate that they have sufficient bandwidth as well as disk and processor capacity to support the deployment in their location.

If not, upgrade the servers. Also, ensure that spare components are available in case of device failure during the deployment.
Because the deployment will occur on the basis of a final release of the deployment tools and scripts, the operating system images, and the application installation packages, look to a replication method to automatically copy these components from a central location to all the deployment servers, including the portable locations.

Using a replication tool, such as DFS Replication Service, that controls bandwidth usage and replication schedules greatly minimizes the risk of not having the proper technologies in place to support the deployment when it occurs. If a replication technology exists, verify that it can support the replication of the deployment components. If no replication technology is in place, plan to deploy one to reduce risk during deployment.
Identify whether each location that includes computers has a deployment server.

If not, design an alternate delivery mechanism. This mechanism can be as simple as having remote users bring their computers in to a central location or as complex as preparing portable deployment servers that can be hooked up to a remote site network to perform the deployment.
After all these components have been addressed, proceed with deployment testing.

Make sure that the test environments replicate all situations the team will encounter during the production deployment. Doing so greatly reduces the risk of discovering an unforeseen situation during the production deployment.

Appendix B: Performing Infrastructure Remediation

Additional resources