Browse Prior Art Database

Pattern-based software detection Disclosure Number: IPCOM000252647D
Publication Date: 2018-Jan-31
Document File: 1 page(s) / 17K

Publishing Venue

The Prior Art Database

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 67% of the total text.

Software pricing has become a very important factor in running a business. We are also observing a shift from perpetual license models to subscriptions and usage-based licensing. To achieve return on investment vendors are introducing more aggressive practice of auditing the customers. To avoid non-compliance, companies need to rely heavily on software asset management tools, which are not always able to detect all instances of software running in infrastructures in time (due to limited access, tool not being deployed everywhere, etc.) Our idea is to use machine learning techniques to suggest existence of undetected software and correct historical data.

After receiving a change, the system monitors if all machines in the group correctly reports the change – if not, we suggest that the change took place and ask for confirmation. Computer groups can either be determined automatically (based on IP domain, hostnames, installed software, etc.) or manually determined. For example, a group of 5 computers is monitored. If a change occurs, the algorithm checks if all members of the group reflected the change, if not a suggestion to re- check the machine occurs.

Use case:

An infrastructure is being monitored by an asset management tool.

Based on the currently detected application (identical or almost identical set of software)

and history of changes (the correlation of software appearance/disappearance between

the computers) + additional optional factors (hostnames, OSes, ip addresses...