I have been remiss in my blog writing (nothing new there) – but some fairly amazing numbers came out from an audit I performed last week. They highlight some of the dangers in manual audit/inventory work and I thought it was blog worthy…

To set the scene: Large european government department. Circa 70,000 users. Oracle wants to audit said department and has been trying for over a year to get an accurate picture of the deployments of their software in order to accurately negotiate a deal.

The department has hundreds of locations spread across a country. Lots of data centers, lots of servers, little time to spend gathering audit information… BUT – they spend some time, in one particular data center over a 3 month period they audit circa 200 servers to complete an Oracle Server Worksheet (OSW) – and they find 15 machines running Oracle.

The person running the license program finds out about iQuate, thinks it could be a good idea to validate the findings and we spend 1 day on site scanning the same 200 machines in the same data center.

The results are pretty astonishing. Manual audit finds 15 machines with Oracle installed, all running. We find 22 (3 of these had Oracle installed, but it wasn’t running, so could be removed). Manual audit says 42 processors (all bar 1 processor noted as single core) on the machines running Oracle, we find 28 processors (all bar 4 of which were actually multi-core…). The manual audit says €1,200,00 worth of Oracle software deployed. The automated audit (which found more machines!) says €800,000 worth of Oracle running on the machines.

€400,000 savings in a single data center, representing about 10% of the total server estate – within a proof-of-concept framework (ie: VERY minimal cost to the organization in question!).

Obviously, we are engaging with the government in question to extend the scope of this audit to 100%, rather than just 10% of the estate. The savings should be massive, the effort in deployment minimal. Speed, accuracy and cost savings – blog worthy I thought :)

JK