<- Back to laws

Parallel computing limit model

Amdahl's Law

The total speedup from parallel hardware is capped by the fraction of work that remains serial.

Scientific statusMathematical performance model
Predictive formBounded speedup
DomainParallel systems
EvidenceModel + benchmarks
Key limitationFixed workload assumption
Common misuseParallelism is pointless
INTERACTIVE MODEL

S(N) = 1 / ((1 - p) + p / N)

N is the number of processors or workers; p is the parallelizable fraction. The serial fraction 1 - p becomes the limiting floor.

Run the same workload against one worker and an adjustable worker pool. The race follows Amdahl's ideal fixed-size model; communication, synchronization, bandwidth, imbalance, and changing problem size remain outside it.

6.4Fixed-workload model speedup
(x)
0 %99 %
SERIAL-BOTTLENECK RACEParallel lanes multiply. The unsplittable gate does not.
Interactive visual model for Amdahl's Law.
MODEL SPEEDUP0.0xINFINITE-WORKER CEILING0.0x

Both racers process the same fixed workload. More workers compress only the parallel section; the coral serial gate fixes the asymptotic ceiling.

CHANGE
Parallelizable fraction
WATCH
speedup
MEANING
Run the same workload against one worker and an adjustable worker pool. The race follows Amdahl's ideal fixed-size model; communication, synchronization, bandwidth, imbalance, and changing problem size remain outside it.
VISUAL MODEL

The serial tail becomes the ceiling.

As workers increase, the parallel part shrinks, but the serial portion stays. Eventually extra workers wait on the part that cannot be split.

serialparallel workersspeedup ceiling
01 / MEANING

What it actually says

Amdahl's Law is a deliberately sharp model: if a fixed workload has a fraction that cannot be parallelized, that serial fraction limits total speedup no matter how many processors are added.

The point is not that parallel computing is futile. The point is that system speed is governed by the whole critical path: serial work, coordination, memory movement, scheduling, and the portion of the workload that actually scales.

Compact formS(N) = 1 / ((1 - p) + p / N)
Best interpretationParallel systems evidence in computing.
Important cautionFixed workload assumption.
"A useful law compresses a pattern. It does not erase the conditions that make the pattern true."
02 / ORIGIN

How the idea developed

The modern form emerged through observation, argument, and later refinement. The timeline separates the first insight from the version now used in textbooks and practice.[1]

19671967

Gene Amdahl presents the fixed-workload argument against simplistic massively parallel expectations.

19881988

Gustafson reframes the issue for scaled workloads, where larger problems use more processors productively.

1990s-2000s1990s-2000s

Multicore and cluster computing make parallel speedup a mainstream engineering concern.

TodayToday

Accelerators, distributed systems, and AI workloads require both Amdahl-style bottleneck analysis and scaled-system thinking.

Historical cautionEponymous laws often change after their first publication. Popular wording may be broader and cleaner than the original evidence.
03 / MECHANISM

How the pattern works

The relation becomes useful only when its mechanism, measurement process, and operating range are visible.

01Serial fraction

Any work that must happen in sequence remains on the critical path.

02Diminishing returns

As N grows, p/N approaches zero, leaving 1 - p as the floor.

03Overheads

Real systems add communication, synchronization, memory, and scheduling costs beyond the ideal model.

MODELS(N) = 1 / ((1 - p) + p / N)

N is the number of processors or workers; p is the parallelizable fraction. The serial fraction 1 - p becomes the limiting floor.

04 / APPLICATIONS

Where it earns its keep

Applications are strongest when the law changes a decision, measurement, model, or experiment rather than merely providing an analogy.

PERFORMANCE ENGINEERING

Find the bottleneck

Application

Profiling should identify the serial or contended part before buying more hardware.

PROFESSIONAL NOTE

Optimize the critical path, not the most visible code.

SYSTEM DESIGN

Choose architecture

Application

Amdahl-style analysis clarifies when parallel workers, caching, batching, or algorithm changes will matter.

PROFESSIONAL NOTE

Sometimes changing the algorithm beats adding cores.

PROJECT PLANNING

Set speedup expectations

Application

Teams can avoid promising linear gains from hardware expansion.

PROFESSIONAL NOTE

A speedup claim should state workload, p, N, and overhead assumptions.

05 / LIMITS & MISUSE

Where it stops working

The classic formula assumes a fixed-size problem and an idealized split between serial and parallel work. In many scientific and data workloads, users increase the problem size as hardware grows, which is closer to Gustafson's Law.

The model also omits memory hierarchy, communication topology, I/O, contention, stragglers, failures, and queueing effects, each of which can dominate real performance.

Misuse

"Parallelism cannot help"

Better: It can help enormously when p is high and overheads are controlled.
Misuse

"The serial fraction is known"

Better: It often changes after profiling, algorithm redesign, caching, or batching.
Misuse

"Hardware scaling is linear"

Better: Amdahl explains why linear scaling is exceptional, not default.
Misuse

"Gustafson disproves Amdahl"

Better: They answer different workload questions.
07 / REFERENCES

Sources and further reading

Original publications and serious secondary scholarship are prioritized over summaries.

  1. Gene Amdahl - Validity of the Single Processor Approach to Achieving Large Scale Computing CapabilitiesThe 1967 paper that introduced the classic argument.https://doi.org/10.1145/1465482.1465560
  2. John Gustafson - Reevaluating Amdahl's LawThe scaled-workload counterpoint now known as Gustafson's Law.https://doi.org/10.1145/42411.42415
  3. Lawrence Livermore National Laboratory - Introduction to Parallel ComputingPractical HPC tutorial covering speedup, scalability, and overhead.https://hpc.llnl.gov/documentation/tutorials/introduction-parallel-computing-tutorial
  4. Hennessy and Patterson - Computer Architecture: A Quantitative ApproachStandard architecture reference for quantitative performance reasoning.https://www.elsevier.com/books/computer-architecture/hennessy/978-0-12-811905-1
CONTINUE EXPLORING

Related laws, with the relationship made explicit.

These are editorial connections, not claims that the laws are mathematically equivalent.

CONTINUE READING

Place this law inside the collection.

LAW 020 / 100 PUBLISHED