There are four key areas to consider that can affect the outcome and cost of a data center audit: Initial Data Collection, Data Accuracy, Tracking Changes and Data Audits.
For this article series, we will assume a typical mid-size data center facility containing 200 racks with 20 assets per rack for a total of 4,000 total data center assets.
Initial Data Collection – The typical manual collection costs come to $15 per asset for “readily available” data which includes manufacturer, model, serial number, name, and location.
Data Accuracy – An industry report from Digital Realty Trust showed manual collection results in a 85-90 percent accuracy rate. This also results in a 10-15 percent error rate, or 400-600 assets with incorrect data in our assumption.
Tracking Changes – We make three assumptions here about data center changes. First, much of outages are due to changes. Second, 80 percent of the Mean Time To Repair (MTTR) is based on trying to locate the asset to change. And finally, third, manual change tracking does not track unauthorized changes. The Digital Realty Trust report showed that 74 percent of data center managers were unable to locate a specific server within minutes and 20 percent were unable to locate the asset within a day.
Data Audits – There is a similar cost and effort involved with data audits as with initial data collection efforts.
There are two approaches to IT and data center audits: manual data center audits or implementing an RFID asset tracking solution.
The first has a high cost associated with it. Based on the above factors and assumptions, it is not unreasonable for data center operators to see costs of up to $60,000 or 20 man weeks around manual data center audits. And, these audits are proven to be inaccurate.
By contracts, the current cost of implementing a typical passive RFID solution in this scenario, our typical data center would see a return on investment in the reduced cost of IT audits in fewer than 12 months.