Throughout my career, I have travelled extensively to meet with customers. In preparation for a trip, I pack a bag with my personal belongings and laptop. Unfortunately, throughout the years, I have managed to forget something more often than I should. In most cases when I forget an item, I have to purchase a replacement item, which is an unnecessary cost.
To reduce the risk of forgetting something before I travel, I decided to create a checklist of items I should pack when traveling. First, I determine my requirements, which includes business attire, such as suit, tie, and matching shoes. Then, I create a checklist that I use each time I pack my bag. Checklists assure that all requirements are met. For me, that means that I do not leave until my checklist has been verified against the items in my bag.
Checklists are often used by IT organizations when they are looking to implement a software and/or a service solution. These checklists are often referred to as a Request for Proposal (RFP) or a Request for Information (RFI). They are typically provided to vendors who are trying to earn the organization’s business. Checklists are simply requirements; however, if a list of requirements is not complete, the desired solution will not be completed as expected.
Managing and optimizing software licenses in the data center can be difficult due to complicated licensing models used by many software vendors. Because software licenses are not properly managed by many organizations, they can incur unexpected costs and fines because of non-compliance to their license contracts. As a result, many organizations are now looking to implement a software asset management (SAM) solution as well as with software license optimization capabilities.
A key component of SAM and software license optimization is discovery services. Discovery is a discipline in itself, and is not limited to just scanning servers and software. Discovery means finding all the required data sources, including all contracts, schedules, and entitlements to effectively calculate your license position.
*Be sure to spend time researching your discovery requirements before you start evaluating products.
Before choosing an IT asset discovery solution, it is important to define your discovery requirements so that you can create a checklist; much like creating a checklist when planning a trip. I have met with organizations that have purchased a software solution, only to realize later that the solution did not meet all their requirements. As a result, additional software or services were required to fill the gaps.
Understand How Software Licenses are Calculated Before Determining Discovery Requirements
License metrics give organizations clarity about the liability of each application; however, calculating license metrics can be complicated. License metrics are typically calculated using one of the following type of licenses:
- Processor-Based – includes Cores and Processor Value Units (PVU’s)
- Resource-Based – a count of the resources.
These license variations will require your discovery services to do more than just scan for software and servers. To collect the inventory information required to determine your license position, you might need to discover how many users are using an application, or the number of resources consumed.
Below are 13 requirements that should be considered when listing your discovery requirements.
Data Center Discovery Requirements
1. Scanner must discover the following Operating Systems:
b. Linux, all versions IE, HPUX, Red Hat, Centos etc.
c. AIX to cover all IBM devices
d. Solaris for Oracle
2. Scanner must identify:
a. The unique ID of the server
b. Type of server
*This is imperative for discovering how the servers are linked to each other for Dependency Mapping.
3. Scanner must identify the server processor:
4. Scanner must identify ‘Cores’ associated with the processor
*Required to identify the ‘Core Factors’ used by Microsoft, IBM, Oracle, and several other vendors to calculate licenses.
5. Scanner must identify the number of available ‘Sockets’
6. Scanner must identify the number of available ‘Threads’
7. Scanner must identify links between servers residing in a cluster
*Many organizations are using virtual environments. Vendors typically have specific models for the virtual environment. Most organizations use Hypervisors on Windows and Linux servers. Getting this wrong could create a massive financial risk to the business. Be sure to understand your clustering architecture and the licenses you are entitled to use.
8. Scanner must identify and differentiate different types of partitions for IBM AIX servers
*IBM AIX servers divide their own servers for virtualization using their own naming convention called ‘Partitioning.’ Changes made to partitions can affect IBM licensing models
a. Scanner must discover Fixed partitions
b. Scanner must discover Logical partitions
*Fixed or Logical partitions can have shared or dedicated pools which can have capped or uncapped virtual processors that are assigned to the partitions. Calculating the entitled capacity and the maximum capacity of each partition needs to be performed to understand how the partition is utilized.
9. Scanner must discover Logs on servers
10. Scanner must identify software applications installed and running on every server:
a. Scanner must identify Physical servers
b. Scanner must identify Virtual servers
c. Scanner must identify Cloud servers
11. Scanner must identify how applications are linked to other applications
12. Scanner must identify how applications are linked to services running on the server
13. Scanner must identify how applications are linked to databases
*How the applications are linked will have an impact on how the applications are licensed. For example, a database might not need a license if it is connected to an application which already includes a license.