Granularity

From Wikipedia, the free encyclopedia

See also: grain and film grain (disambiguation)

Granularity is a measure of the size of the components, or descriptions of components, that make up a system. Systems of, or description in terms of, large components are called coarse-grained, and systems of small components are called fine-grained; here coarse and fine are descriptions of the granularity of the system, or the granularity of description of the system.

An example of increasingly fine granularity: a list of nations in the United Nations, a list of all states/provinces in those nations, a list of all counties in those states, etc.

1 In physics
2 In computing
3 In Reconfigurable Computing and Supercomputing
4 In credit portfolio risk management
5 See also

[edit] In physics

A fine-grained description of a system is a detailed, low-level model of it. A coarse-grained description is a model where some of this fine detail has been smoothed over or averaged out. The replacement of a fine-grained description with a lower-resolution coarse-grained model is called coarse graining. (See for example the view of the second law of thermodynamics in the article Maximum entropy thermodynamics)

[edit] In computing

In parallel computing, granularity means the amount of computation in relation to communication, i.e., the ratio of computation to the amount of communication.

Fine-grained, or "tightly coupled", parallelism means individual tasks are relatively small in terms of code size and execution time. The data are transferred among processors frequently in amounts of one or a few memory words. Coarse-grained, or "loosely coupled", is the opposite: data are communicated infrequently, after larger amounts of computation.

The smaller the granularity, the greater the potential for parallelism and hence speed-up, but the greater the overheads of synchronization and communication. (The last two paragraphs are based on FOLDOC.)

[edit] In Reconfigurable Computing and Supercomputing

In Reconfigurable Computing and in Supercomputing these terms refer to the data path width. The use of about one bit wide processing elements like the configurable logic blocks (CLBs) in an FPGA is called Fine-grained computing or Fine-grained reconfigurability, whereas using wide data paths, such as, for instance 32 bits wide resources, like microprocessor CPUs or data-stream-driven data path units (DPUs) like in a reconfigurable datapath array (rDPA) is called Coarse-grained computing Coarse-grained reconfigurability.

[edit] In credit portfolio risk management

In credit portfolio risk modeling, granularity refers to the number of the exposures in the portfolio. The higher the granularity, the more positions are in a credit portfolio, providing a higher degree of size diversification, which in turn reduces concentration risk. This is colloquially known as "not putting all your eggs in one basket".