Whites Papers

The concept of IT organizations sharing statistics through an industry wide database was pioneered by RPI to great success in a measurement facility known as Reliability Plus or R+, from 1970 to 1999. For all of its success, R+ remains until today the only example of this kind of consistent long term data sharing among large numbers of IT organizations that was focused on achieving a specific management objective. R+ gathered statistics monthly from approximately 2,000 large data centers throughout the industry and provided each participating organization with monthly reports on the reliability of all mainframe hardware. What R+ proved in 1970 thru 1999 was that shared statistics achieved the power necessary to accelerate the growth of reliability far beyond what would have happened otherwise. The following are some of the reasons why data sharing is far more effective than relying only on internal statistics:
 

Shared statistics are available sooner. Hundreds of IT organizations can generate a statistically significant measurement in one month that would take one organization hundreds of months to generate.

Shared statistics report on all components. One organization can only monitor the components it has. Shared monitoring allows each IT organization to see the stability of all components used in the industry, or at least all those components used by those IT organizations participating in R+2.

Shared statistics are not easily dismissed by the vendors. Vendors can easily dismiss the data collected by a single IT organization, regardless of its size, as being atypical of the industry as a whole. Shared statistics collected from hundreds of mid to large size organizations throughout the industry are more difficult to dismiss as being atypical.
 

Shared statistics have more clout in the marketplace. If one of the purposes of measuring stability is to get the vendors to aggressively compete in their products' ability to maximize network stability, then it is reasonable to assume that hundreds of organizations collecting and using the same statistics will have significantly more clout than will just one.

Shared statistics send a loud message. That message is that the stability of the network matters. When hundreds of senior IT management join together to measure the stability of their networks that message is heard by the IT workers who build and protect those networks, by corporate workers who depend on those networks, by top corporate executives who are ultimately responsible for those networks, and by vendors whose products and services impact the stability of those networks. When IT organizations measure the stability of their networks using R+2 that stability will become far more stable because no vendor organization wants to be known as the company that undermines that stability and no IT manager or corporate manager wants to find out that its corporate network is one of the least stable in the industry. R+2's measurement not only inform, they motivate those that can improve network stability, to do so.