|This article does not cite any references or sources. (December 2009)|
In computing, the term bucket can have several meanings. It is used both as a live metaphor, and as a generally accepted technical term in some specialised areas. A bucket is most commonly a type of data buffer or a type of document in which data is divided into regions.
Features of a bucket
Various usages relate to different features. There is no usage that is consistent with every one of these features.
- The contents of a bucket are unsorted.
- A bucket has a fixed size, which is determined when it is created.
- A bucket has a limited number of states:
- It may be empty.
- It may be non-empty. Some usages further distinguish:
- It may be partly full.
- It may be full. Some usages further distinguish:
- It may be exactly full.
- It may be overflowing.
Types of bucket
It has been commonly suggested that this usage dates from the days when punched cards and/or punched tape were common. Card and paper-tape punches each had a receptacle for the 'chips' punched from the cards or tape to create the holes (rectangular in a card, round in a paper tape), which often looked like a bucket and was known by this name. By analogy it was jokingly suggested that the CPU, which in those days was an equipment rack of similar size to a large peripheral unit, contained a similar receptacle for lost data.
A bucket is a document of no definite size to which information of interest is added with no structure. Many software packages have a README file which is a bucket containing the very latest information. In IBM culture, such a file is known as a bucket and is opened for critical fixes and fix packages.
Buffers known as buckets are used:
- In the bucket sort algorithm.
- In bucket elimination algorithms.
- In the token bucket (TB) algorithm for network traffic shaping.
- In the leaky bucket (LB) algorithm for network traffic shaping.
- In bucket brigade algorithms.
|This computer storage–related article is a stub. You can help Wikipedia by expanding it.|