Optimized license management is a necessity for all software licenses owned by the enterprise, from desktop to datacenter and beyond. While many organizations understand their license position for the desktop estate, the reality is that licensing in the datacenter still presents a daunting set of challenges that require a robust, automated Software License Optimization solution. Compounding this, it is difficult to take the same set of processes, technologies, and remediation techniques used to manage desktop licenses and apply them to datacenter license management. Organizations need a solution to address the unique license management requirements of enterprise IT environments including the desktop, datacenter and private and/or hybrid clouds.
Applications in the datacenter are strategic and run the business, but they are also typically the most expensive. Optimizing license and maintenance spend in the datacenter typically represents the greatest potential costsavings in the software portfolio. Decreasing ongoing costs of multi-million dollar applications in the datacenter is a fundamental component of overall IT spend reduction. As organizations continue datacenter consolidation through virtualization, license optimization in the datacenter becomes even more critical to reduce the associated risk of software license noncompliance – and more difficult without an automated solution. Datacenter license management challenges include:
- Heterogeneous / Multi-platform environments
- Complex License Models
- License Entitlements: Product Use Rights
Heterogeneous / Multi-platform Environment
In most companies, the datacenter and server infrastructure represent the most diverse computing environment in the company. Companies typically have multiple flavors of Linux, UNIX and Windows operating systems in the estate. Effective license management within this heterogeneous environment requires a robust multi-platform discovery and inventory solution.
The license optimization solution should also leverage existing IT investments by integrating with common discovery and inventory tools used in the datacenter, such as BMC Atrium Discovery and Dependency Mapping (ADDM), IBM Tivoli, HP Discovery and Dependency Mapping Inventory (DDMI), etc. An advantage of this approach is that yet another agent doesn’t need to be deployed. Once the hardware and software inventory data is collected, it must be processed and normalized to quickly identify and categorize server-based applications, including publisher, title, version and edition.
Agent and Agentless Inventory
For organizations that do not have a standard inventory solution for the server estate the datacenter license management solution should provide a multi-platform agentless and/or agent-based inventory capability. Particularly for datacenter server environments, an agentless inventory capability is often required for server performance and security reasons—an agent may not be allowed to be installed on many of these systems. As shown in the pie chart above, the typical datacenter has a mix UNIX, Linux and Windows operating systems that may include: Sun Solaris, HP UX, IBM AIX, RedHat Linux, CentOS, SuSE Linux, Fedora, Windows Server, and other OS platforms.
In some cases, specialized discovery and inventory tool capabilities are needed. This is true for Oracle and Microsoft SQL Server database software, for example. The Oracle database products require the collection of important option, and management pack information, and Microsoft SQL Server requires data to determine the installed edition. This information is not available in most standard inventory tools but is necessary for license management.
Virtual Machine Discovery & Inventory
With the growing deployment of server virtualization technologies from vendors such as VMware (ESX/vSphere) and Microsoft (Hyper-V), organizations must also collect inventory for virtual machines across the network. In these environments, it’s necessary to not only bring back inventory of software on each virtual machine, but also identify the relationship between virtual machines and physical host servers. The hardware details— number and type of processors and cores, memory size, etc., are important, as is the allocation of hardware resources to virtual machines. This data is often required to perform license management in virtual environments, particularly where the license model is based on server and/or processor characteristics. Examples of vendors that use processor/core based licensing for enterprise server software abound: Microsoft, IBM, Oracle, and Symantec, to name just a few.
Complex License Models
License metrics in the datacenter are becoming increasingly complex. More publishers are moving to capacity-based licensing models which make it difficult to calculate an accurate license position without automation. Organizations must be able to determine how many licenses are consumed based on the license models in effect and the capacity characteristics of installed machines (e.g. processors, cores, and memory). As a simple example, the organization must calculate processor-based license consumption based on the number of processors in each (physical or virtual) server to determine whether the appropriate number of licenses have been purchased and allocated in the most optimal way.