Concept of Temporal Parallelism
In order to make clear what is meant by parallelism inherent in solution of a problem, let's discuss an example of submission of electricity bills. Assume there are 10000 residents in a locality and they are supposed to put forward their electricity bills in one office.
Let's suppose the steps to submit the bill are as follows:
1) Go to the suitable counter to take a form to submit the bill.
2) Submit the filled form with cash.
3) Get the receipt of submitted bill.
Suppose that there is just one counter with just single office person doing all the tasks of giving application forms, accepting the forms, counting the cash, returning the cash if the need be, and giving the receipts.
This circumstance is an example of sequential execution. Let us suppose the estimated time taken by several of events is as follows:
Giving application form = 5 seconds
Accepting filled application form and counting cash and returning, if required =
5 minutes i.e. 5 ×60= 300 sec.
Giving receipts = 5 seconds.
Total time taken in processing one bill = 5+300+5 = 310 seconds.
Now, if we have 3 persons sitting at 3 different counters with
i) One person giving bill submission form
ii) One person accepting cash and returning, if required and
iii) One person giving the receipt.
The time needed to process one bill would be 300 seconds because the first and third activity will overlap with the second activity which takes 300 sec. while the first and last activity takes only 10 seconds each. This is an illustration of a parallel processing process as here 3 persons work in parallel. As 3 persons work in the same time, it's called temporal parallelism. However, this is not so good example of parallelism in the sense that one of the actions which is the second action takes about 30 times of time taken by each of the other two actions. The word 'temporal' means 'pertaining to time'. Here, a task is divided into many subtasks, and those subtasks are implemented all together in the time domain. In terms of computing application it can be said that parallel computing is possible, if it is possible, to break the problem or computation in to equal independent computation. Preferably for parallel processing, the task must be divisible into some activities, each of which takes roughly equal amount of time as other activities.