Amdahl's law
E219378
Amdahl's law is a formula in computer architecture and parallel computing that predicts the maximum performance improvement achievable by parallelizing parts of a system, given that some portion must remain serial.
All labels observed (3)
| Label | Occurrences |
|---|---|
| Amdahl's law canonical | 5 |
| Amdahl's argument | 1 |
| Amdahl’s law | 1 |
How this entity was disambiguated
This entity first appeared as the object of triple T1961130 — resolving that mention is where its identity was fixed. The disambiguator weighed these candidate entities and picked the highlighted one (or “None”, minting a new entity). This is how homonymy is resolved: the same surface form can point to different entities.
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Amdahl's law Context triple: [Gene Amdahl, knownFor, Amdahl's law]
-
A.
Moore's law
Moore's law is an observation and prediction that the number of transistors on an integrated circuit—and thus computing power—tends to roughly double at regular intervals, driving exponential growth in digital technology.
-
B.
Wirth’s law
Wirth’s law is the observation that software tends to become slower more quickly than hardware becomes faster, often negating the benefits of improved computing performance.
-
C.
"Computer Architecture: A Quantitative Approach"
"Computer Architecture: A Quantitative Approach" is a seminal textbook that rigorously explores modern computer architecture design and performance analysis, widely used in academia and industry as a definitive reference.
-
D.
"How to Make a Multiprocessor Computer That Correctly Executes Multiprocess Programs"
"How to Make a Multiprocessor Computer That Correctly Executes Multiprocess Programs" is a seminal paper by Leslie Lamport that introduced foundational concepts for ensuring correctness and consistency in concurrent and multiprocessor systems.
-
E.
Kluge's law
Kluge's law is a proposed sound law in Proto-Germanic historical linguistics that explains the development of certain geminate consonants from earlier consonant clusters.
- F. None of above. chosen
- G. Unsure - the case is ambiguous/there is not enough information to decide.
NED2
Entity disambiguation (via description)
gpt-5-mini-2025-08-07
Target entity: Amdahl's law Target entity description: Amdahl's law is a formula in computer architecture and parallel computing that predicts the maximum performance improvement achievable by parallelizing parts of a system, given that some portion must remain serial.
-
A.
Moore's law
Moore's law is an observation and prediction that the number of transistors on an integrated circuit—and thus computing power—tends to roughly double at regular intervals, driving exponential growth in digital technology.
-
B.
Wirth’s law
Wirth’s law is the observation that software tends to become slower more quickly than hardware becomes faster, often negating the benefits of improved computing performance.
-
C.
"Computer Architecture: A Quantitative Approach"
"Computer Architecture: A Quantitative Approach" is a seminal textbook that rigorously explores modern computer architecture design and performance analysis, widely used in academia and industry as a definitive reference.
-
D.
"How to Make a Multiprocessor Computer That Correctly Executes Multiprocess Programs"
"How to Make a Multiprocessor Computer That Correctly Executes Multiprocess Programs" is a seminal paper by Leslie Lamport that introduced foundational concepts for ensuring correctness and consistency in concurrent and multiprocessor systems.
-
E.
Kluge's law
Kluge's law is a proposed sound law in Proto-Germanic historical linguistics that explains the development of certain geminate consonants from earlier consonant clusters.
- F. None of above. chosen
Statements (48)
| Predicate | Object |
|---|---|
| instanceOf |
computing law
ⓘ
formula ⓘ performance law ⓘ theorem in parallel computing ⓘ |
| alsoKnownAs |
Amdahl's law
ⓘ
surface form:
Amdahl's argument
|
| appliesTo |
GPU acceleration
ⓘ
I/O-bound and CPU-bound workload decomposition ⓘ distributed systems ⓘ multi-core processors ⓘ |
| assumes |
fixed total problem size
ⓘ
no overhead for communication and synchronization in its basic form ⓘ remaining portion of the workload is perfectly parallelizable ⓘ some portion of the workload is strictly serial ⓘ |
| contrastedWith | Gustafson's law ⓘ |
| coreConcept |
diminishing returns from adding more processors
ⓘ
serial bottleneck limits overall speedup ⓘ speedup is bounded by non-parallelizable fraction ⓘ |
| defines |
N as number of processors or parallel units
ⓘ
P as fraction of workload that is parallelizable ⓘ S as overall speedup ⓘ |
| describes |
effect of serial fraction on parallel speedup
ⓘ
limit of performance improvement from parallelization ⓘ maximum speedup of a system using multiple processors ⓘ |
| field |
computer architecture
ⓘ
high-performance computing ⓘ parallel computing ⓘ performance analysis ⓘ |
| formalExpression | S = 1 / ((1 − P) + P / N) ⓘ |
| historicalContext |
formulated in the 1960s
ⓘ
introduced in the context of mainframe and multiprocessor design ⓘ |
| implies |
even small serial fractions severely limit speedup
ⓘ
improving serial portion can be more beneficial than adding processors ⓘ maximum speedup as N approaches infinity is 1 / (1 − P) ⓘ |
| influenced |
design of parallel architectures
ⓘ
thinking about limits of parallel processing ⓘ |
| limitation |
assumes constant serial and parallel fractions independent of N
ⓘ
does not account for changing problem size ⓘ ignores communication and synchronization overheads in simple form ⓘ |
| namedAfter | Gene Amdahl ⓘ |
| relatedTo |
bottleneck analysis
ⓘ
law of diminishing returns ⓘ parallel speedup ⓘ strong scaling ⓘ |
| usedFor |
analyzing trade-offs in system optimization
ⓘ
estimating scalability of parallel programs ⓘ evaluating benefits of hardware acceleration ⓘ guiding design of multiprocessor systems ⓘ performance modeling ⓘ |
How these facts were elicited
The pipeline generated the facts above by prompting gpt-5.1 with this entity's name + description and the instruction below.
Instruction
You are a knowledge base construction expert. Given a subject entity and a description of it, return factual statements that you know for the subject as a JSON list of dictionaries(triples), where keys must be "subject", "predicate" and "object". The number of facts may be very high, between 25 to 50 or more, for very popular subjects. For less popular subjects, the number of facts can be very low, like 5 or 10. # Requirements - If you don't know the subject at all, return an empty list. - If the subject is not a named entity, return an empty list. - Include at least one triple where predicate is "instanceOf". - Do not get too wordy. - Separate several objects into multiple triples with one object.
Input
Subject: Amdahl's law Description of subject: Amdahl's law is a formula in computer architecture and parallel computing that predicts the maximum performance improvement achievable by parallelizing parts of a system, given that some portion must remain serial.
Referenced by (7)
Full triples — surface form annotated when it differs from this entity's canonical label.
this entity surface form:
Amdahl’s law
subject surface form:
Gene Amdahl
this entity surface form:
Amdahl's argument
subject surface form:
Gene Amdahl