HI Before going into the explanation of seed value I have to say how the testers faced the problems with out using the seed values
Here a small example is my friend
He is a tester in one of the company, he collected some data to test one of the application module after testing is over, he moves to another module, at that time he thinks that he has to do the testing by combining the old data with new data, if he did not saves that data he has to prepare again or if he saves no problem but its take time and takes more memory to save all data of modules, he faces some problems while using that old data
BY this you got some idea what is the use of seed value, any way I can give clear information about this in detail
A seed is a significant number on which most Random Data Generation Algorithms base upon. Software Testing benefits from seed values from the fact that they allow re-creation of old random data exactly in the same manner when needed.
Let us say a tool in TestersDesk.com that provides random data along with a seed was used to generate 100 values. For the sake of understanding a seed, let us say the 67th value from the generated 100 values has detected a bug in the application under test. Some time later, when there is a new build on which you need to test with both newly created random data and old data which detected a bug; there are two choices- either store the old data somewhere and reuse it or just provide the old seed value and recreate exactly the same data again from TestersDesk.com.
As a good practice, it might be worthwhile to store the seed values from the generated random data instead of storing the random data itself (which may eventually lead to a large un-maintainable repository).
There might be some disadvantages also, frankly speaking I don’t know the disadvantages, if anybody knows please post the comments or otherwise you can send to my mail