Advanced Computer architecture

Amdahl’s Law
Lecture 3
 Speedup
• the speedup factor of a parallel system as the ratio
between the time taken by a single processor to
solve a given problem instance to the time taken by a
parallel system consisting of n processors to solve the
same problem instance.

𝑇(1)
𝑆 𝑛 =
𝑇 𝑛
T(1):Serial time
T(n): Parallel time
 Amdahl’s Law
• Analyze whether a program merits parallelization
• Amdahl’s law assumes that the problem size is
fixed. It provides an upper bound on the speedup
achievable by applying a certain number of
processors.
𝑛
𝑆 𝑛 =
1+ 𝑛−1 𝑓
n = No. of processors
f: sequential fraction,
where
0≤𝑓≤1
 Amdahl’s Law Example
If 90% of the computation can be parallelized, what is the
maximum speedup achievable using 8 processors?

Solution:
f=100%-90%=10%=0.1
𝑛
𝑆 𝑛 =
1+ 𝑛−1 𝑓

8
𝑆 𝑛 =
1 + 8 − 1 ∗ 0.1

𝑆 𝑛 ≈ 4.7