Oracle Real Application Clusters

Parallel processing is the use of multiple processors to execute different parts of the same program simultaneously.

The main goal of parallel processing is to reduce wall-clock time. Both scenarios discussed in the previous section are examples of parallel processing and both illustrate the goal of distributed processing: the reduction of wait time to solution achievement.

Within the example of the superhighway above, the power and weakness of the parallel approach become visible by taking it to the extreme. In other words, as the number of lanes on the highway is increased, there will be a characteristic speedup curve, demonstrating how, up to a certain number of lanes, it is beneficial. However, anything over a certain point probably will give a reduced or negative benefit.

Figure 6.1 represents the increase in the number of lanes to the movement of traffic on a highway. It becomes evident that after a point of six lanes and a speed limit of 70 mph, the curve starts going down, showing a decrease in speed. This is another example of the law of diminishing returns that was discussed earlier in Chapter 1. Too much of something may bring back a lesser satisfaction or benefit. In this case, with the increase in lanes, after a certain point the speed of traffic movement reduces. This is because more vehicles end up crossing lanes, which causes a disruption of fast-moving traffic, either in exiting from the highway or in joining the highway. This back and forth movement of vehicles to cross lanes results in reduced traffic speed. A similar behavior will also be noticed in the example related to information technology. Parallel processing is the:

• Division of work into smaller tasks

• Assignment of many smaller tasks into multiple threads to execute simultaneously

• Reduction of wall-clock time for execution of computer programs

Figure 6.1: Parallel proces sing example.