In our last post, we discussed several critical and difficult-to-gather data variables needed to answer the question, “Who is using what application and for how long?" The first variable we’ll look at is “who is the user.”
Users are easy to identify as employees. Their data is in the corporate HR database. They are easy to spot sitting at their desk, walking around the office, or meeting in a conference room. Identifying whether or not they are using a particular application at a specific time is not as easy. Making the connection between the user and the application requires having mechanisms to identify users and applications, as well as the ability to tie them together.
To identify the user, we need to know when she logs into a device, such as a network workstation. Logging in identifies who the user is, as well as the user’s time and location. Login information doesn’t specifically tell us anything else about “who” the user is, such as her title, the group she’s with, what applications she’s allowed to use, etc. This information is available, but is typically via other applications or databases.
We’ll dig into the “what the application” is in our next post, but this information is important to mention now, because if we can’t track the use of a specific application, then we can’t associate the user. Problematic applications for use association are SaaS applications, standalone desktop applications, or applications that obtain their licenses from something other than a license server, because each offers potentially trackable information from different sources.
Associating who the application user is requires either:
A monitoring solution capable of gathering information about who the user is, as well as application information from multiple data sources,
Cobbling together a collection of tools that each address a sub-set of the application universe.
Note that these tools don’t necessarily address the “who the user is” question. The reporting solution approaches are:
- Server Log File reader reporting solutions provide information on how many licenses are checked out at any point in time. These solutions do provide information on “who” the user is, such as the user name, workstation name and IP address, but only for application licenses delivered from a license server. Applications having different license delivery mechanisms require additional solutions. Reporting additional attribute information such as what group the user belongs to may require the organization to create a data mart to combine use and attribute information.
- Software Asset Management reporting solutions track software installations and can identify that a particular machine had a certain application installed on it. They list the last time an application was run. They don’t track when the user logged into a computer, so they are very poor solutions for who the user is at a particular point in time.
- Network monitoring solutions monitor network traffic, but they won’t necessarily track or identify who a user is on a particular machine. They don’t track applications running on a machine if the license is locally installed. They are great for tracking network traffic, but not for monitoring desktop applications.
- Data center-centric applications that monitor the number of virtual machines, or data base instances, or CPUs will provide information on aspects of server workloads, but have no notion of an individual workstation, who a user is, etc.
- Desktop agents report who the user is when the user logs in. Agents can capture application-specific information such as the application name, version, etc. when it runs locally on the workstation. Agents don’t need to know where the license comes from to capture use information, although that information might be of importance in reporting the “what application” portion of the overall question, which we’ll cover in our next post. In addition, agents can capture when the application is launched, and perhaps whether the application is actively being used. Sophisticated agents may also be able to monitor individual windows. Agent-based systems typically know who the user is because this information is generated during agent installation and updated or validated when the user logs in.
Cetrus Process Meter is an agent-based reporting solution that can answer the “who, what and when” of application usage. If you’d like to learn more, send a note to firstname.lastname@example.org.