Skip to content
This repository has been archived by the owner on Mar 3, 2022. It is now read-only.

Email comment: Belarc, Inc. comments on OMB software policy #7

Open
OMBPublicComments opened this issue Dec 29, 2015 · 0 comments
Open

Comments

@OMBPublicComments
Copy link

It’s great to see your initiative on trying to manage the federal government’s software spend. There is little need to reiterate the ever increasing costs and huge potential for annual savings in this area. We would like to make the following comments:

• The memo mentions using DHS’ CDM tools to report on software inventory and usage. We believe that some Agencies have far more effective tools than the current CDM offerings and that the current CDM offering lacks some important capabilities for software license management, as follows:

o Enterprise wide capability. The software inventory and usage data must be readily available on a near continuous basis from all IT assets covering the enterprise. Many Agencies have many tens of thousands of IT assets across many geographic locations. The system should operate automatically and not rely on local servers and require any local admin time and effort to manage.

o Automatic discovery of complex server software, dependencies and related hardware details. Server software likely constitutes the majority of the federal government’s software spend. Vendors such as Microsoft, Oracle, IBM, SAP license their server software based on complex metrics. In order to determine the licensing position of these products it is necessary not only to discover the installation of the software, but also the dependencies between the virtual machines where the software is installed and the physical host machines; the hardware details such as CPUs, cores, whether hyper-threading is enabled; is the server part of a cluster; and even if the Options have been used or not (for Oracle). The current CDM toolset does not accomplish this.

o Discovery of expensive desktop software. Many government services are dependent on CAD and GIS (geographical information system) software, and these are often very expensive products. In our experience AutoCAD and ESRI are in particular use in the federal government. Your current CDM offerings do not adequately discover these products to enable licensing.

o Discovery and usage of Cloud software. This will likely become an ever increasing part of the government’s software spend. Licensing Cloud products may make counting devices easier, i.e. one user can access up to five devices for example, but tracking whether the user has actually used any of those devices or is no longer with the Agency will be important. The current CDM offering does not have this capability.

• There was little mention of the need to merge the discovered data with the Agency’s purchase records and licensing rules for the software products. This is especially important for complex server software where the licensing metrics are complex and some form of automation should be used to determine a true licensing position. In fact, this should ideally be done on a continual basis, where the discovery and usage data is automatically used to link to the latest purchase records and licensing metrics.

• There is likely a need to bring in outside expertise on negotiation with the major vendors, in particular Microsoft, Oracle, IBM and SAP. These people bring in years of experience, often with the software vendors themselves, to help the purchaser manage the licensing process more effectively.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant