This equation Amdahl's law on basis of computing-centric system which never takes into account the potential cost of data preparation.
Therefore, traditional Amdahl's Law is not fit for heterogeneous computer system, we will take consideration of the overhead of data preparation in asymmetric multicore system.
Despite some criticisms, Amdahl's Law is still relevant as we enter a heterogeneous multi-core computing era.
In Section 4, we discuss a number of recent approaches that extend Amdahl's law for the multicore architectures.
Amdahl's Law  is one of the few fundamental laws of computing that contribute to systems' performance enhancement.
Despite its simplicity, Amdahl's law is applied broadly and gives important insights such as it when f is small, optimization will have little effect.
Contradicting Amdahl's law, equation (2) indicates a linear speedup.
By defining different f(Q), we find that Amdahl's law is the low-extremum of our speedup function and Sandia's is the high-extremum.
Equation (7) also shows that Amdahl's law is still true for certain kinds of problems even though the problem size is increased.
By doing so, we can provide better performance than Amdahl's law computed with typical load balancing (i.
In this target application, the performance gain obtained by typical parallelization is limited by Amdahl's law , as shown in Fig.
The notion that Amdahl's law has somehow been violated is encouraged by an error in Gustafson's paper.
In view of the diversity of problems to which multiprocessors may be applied, there seems to be no single "best method or fixed set of assumptions for characterizing potential efficiency, but Amdahl's law, when correctly applied, continues to provide helpful insight.
REEVALUATING AMDAHL'S LAW At Sandia National Laboratories, we are currently engaged in research involving massively parallel processing.
If N is the number of processors, s is the amount of time spent (by a serial processor) on serial parts of a program, and p is the amount of time spent (by a serial processor) on parts of the program that can be done in parallel, then Amdahl's law says that speedup is given by Speedup = (s + p)/(s + p/N) = 1/(s + p/N), where we have set total time s + p = 1 for algebraic simplicity.