Software licensing models evolved in the 1990’s with the emergence of the floating license model, a model still in wide scale use today, largely in technical software markets. At the time, the floating license model was an evolution of machine-locked software that was prevailing on standalone or lightly-networked personal computers. Interestingly, the floating license model is taking its next evolution, with virtual machines and Cloud Computing. However this time, floating licenses are probably going to do for enterprise software what they did for technical software the first time around.
Floating Licenses 1.0 – License Mobility for Vertical Software
The floating license model evolved to meet the needs of workgroup computing in the 1990’s. In the late 1980’s and early 1990’s the computing environment began to evolve with the availability of the networked workstation from Sun Microsystems and others. In a networked environment, resources such as printers and storage are shared among users in a local area network. A natural evolution was the sharing of a software license – hence the term “floating license”. With the floating license model, when the user starts their software on a machine in the network, that software will automatically check-out a license from a license server (so long as licenses are available). As long as the license is checked-out, the user is able to run the software associated with the license, and no one else can use that license. When the user is done using the software, the license is automatically returned to the license server for others to use. The license server can contain multiple quantities of different licenses, allowing entire workgroups to share licenses for different software titles. If a license isn’t available for a particular user because they are checked-out, the user can wait for the license to be available, and/or, as the system administrator to buy more licenses. The floating license has an interesting attribute – because a software license is sharable, it can be used on a variety of different machines over the course of its life. This means that the software license can’t be tied to a particular machine type as identified by the number of cores, CPU’s, or processors. The floating license model works very well in markets where customers require access to a wide variety of software titles over the course of a day.
Over time, the flexibility of the floating license model and associated license server technology provided the basis for more flexible license models in wide scale use today (such as subscription licensing, time-based “peak usage” licensing; forms of pay-per-use licensing models), along with a variety of asset management and reporting tools associated with the license server technology. This in turn, allows for more efficient usage and tracking of software, and, the capability to construct more sophisticated license agreements with both the consumer and independent software vendor understanding usage information.
Floating Licenses 2.0 – License Mobility for Enterprise Software
Fast forward to a year in the future, say 2011, or 2012. The floating license model is becoming the paradigm for the use and management of enterprise level software, for similar reasons it became popular with technical software.
With virtualization technology, enterprise software licenses previously dedicated to physical machines are increasingly being deployed in unique virtual machines. This virtual machine can be moved to the particular physical machine that is available and best suited to the requirements of the user.
With virtualization moving technology such as VMotion, the situation becomes more interesting. Now users, have the capability to dynamically move those virtual machines from one machine to another, without disruption and without delay. For now, this is a convenient method to provide a high-availability solution for enterprise software. If the machine running the main virtual machine fails, the virtual machine can be “auto-magically” moved to another physical machine.
Now, let’s extend the paradigm. Suppose an enterprise has several enterprise applications running on premise, but they are running out of machine bandwidth, perhaps due to some seasonal variability in the business. Wouldn’t it be convenient to move one or more of the virtual machines to another machine on-premise or, into the cloud and let another hosting center run the application for awhile? Alternatively, suppose an enterprise initially decided to have their purchased enterprise software hosted by a cloud provider initially, but over time, decided to move on-premise to lower costs or gain more control? Wouldn’t it be ideal to simply transfer a virtual machine (including all associated data) from the hosting provider in the cloud to an on-premise machine?
Of course, this is a reach right now with virtual machine moving technology, compliance technology, and security, but it’s not too far-fetched to think we may see it in the next few years.