The outcomes of analysis of Amdahl's law are:

1) To optimize the performance of parallel computers, modified compilers need to be developed which should aim to decrease the number of sequential operations pertaining to the fraction α.

2)  The manufacturers of parallel computers were discouraged from manufacturing large- scale machines having millions of processors.

There is one main shortcoming identified in Amdahl's law. According to Amdahl's law, the workload or the difficulty size is always fixed and the number of sequential operations mainly remains similar. Thus, it assumes that the distribution of number of sequential operations and parallel operations always remains similar.  This situation is shown in Figure and the ratio of sequential and parallel operations are independent of difficulty size.

No. Of processors

Fixed load for Amdahl's Law

Though, practically the number of parallel operations increases according to the size of difficulty.  As the load is assumed to be fixed according to Amdahl's law, the implementation time will keep on decreasing when number of processors is increased. This situation is shown in Figure given below. This is in Introduction to the processes operation.

No. Of processors

Execution Time decreases for Amdahl's Law