Apr

7

Redundancy is one of the keys to digital cell phone transmissions, and packet transmissions for the internet, human speech, credit card numbers, music composition. The list goes on and on, but should include the market. In speech, typically people say the same thing over and over, to guaranty the message gets through. Digital cell phone technology uses some sort of redundant error correction to insure the correct message. Musical composition often has three verses, and repeats the theme to get the message through.

The market does the same. The mechanism is the result of trial and error, to some degree, but also of communication, error correction. A minimum of three is needed to provide some sort of error correction, and to insure transmission of the message. This is why we often see things in threes. It is good to know or expect repetition or redundancy as it gives an edge. For some reason the news and commentators seems to think rather of endless continuation as the normal mode.

Paolo Pezzutti adds:

Redundancy increases reliability of systems, usually in the case of a backup. You can find in many critical-performance systems and applications that some components or modules can be at least doubled. When you have a federated system, for example, you can choose to have a central "intelligent' core and a number of "non-intelligent" sub-systems, or you can have "intelligent" subsystems providing a higher degree of resilience to failures. This is typical of some combat systems on board ships for example. The point is that not only redundancy adds reliability, but it increases also the performance, because intelligence is distributed throughout the system of systems and decentralization is a more efficient and effective solution (there are less bottlenecks and so on).

Redundancy and reliability, however, have a cost. When designing a system you have to weigh costs and benefits to find a balance that meets the user requirements. Markets find dynamically a balance between costs and benefits through the price discovery process. Also in this case, network-enabled players that apply a decentralized approach have an advantage in situational awareness, speed of evaluating the situation and making decisions, and speed of execution have an advantage over bureaucratic, centralized and slow players.

Phil McDonnell comments:

Redundancy can be very good but there are some occasions when it accomplishes less than one might think. For example, most data centers have more than one server. But if they are running on the same electrical power system they are still vulnerable to the loss of that common critical resource.

Another example might be when the sources of failure are not independent. One example using two servers might be if both are plugged into the same wall plug. They are susceptible to common power surges and lightning strikes transmitted over the power lines. Even several computers connected via long network cables can be simultaneously damaged by the EM pulse from a nearby lightning bolt.


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search