speedup

(redirected from speedups)
Also found in: Thesaurus.

speed·up

 (spēd′ŭp′)
n.
1. An increase in speed; acceleration.
2. A required acceleration of work or production without an increase in pay.
ThesaurusAntonymsRelated WordsSynonymsLegend:
Noun1.speedup - the act of acceleratingspeedup - the act of accelerating; increasing the speed
hurrying, speeding, speed - changing location rapidly
References in periodicals archive ?
Speedups of 55 times were reported over non-optimized CPU version.
Consequently, parallelization results are improved giving higher speedups with some benchmarks.
We observe that Shared Input and Output version speedups the code by 13% and 78%.
Their experimental results report 5.6x and 9.7x speedups. The work in [14, 15] exploits a computer cluster system to parallelize CBIR with several computers connected by high-speed network.
The potentials of the maximum of speedups using architecture of a symmetric, asymmetric or dynamic multicore is obtained.
The hardware accelerators proposed achieved different speedups over a traditional general purpose processor.
Our objective is to solve the system of linear equations using computers equipped with either single core or multi-core processor and compare the speedups achieved, as most of the academic and research community tends to use a dedicated computer to solve their problems.
Many of those speedups appear at depths of around 2,900 kilometers, says Jung-Fu Lin, a geophysicist at the Lawrence Livermore (Calif.) National Laboratory.
We reported speedups of as much as 17.5 times on 24 processors of a 32 processor shared memory SGI Origin 2000 (1).
1997; 1999] is a value-specific, dynamic compilation system that has good potential for producing speedups on larger, more complex C programs.
"Assembly-line speedups continue to cause excessive injuries, stifling heat and oppressive working conditions remain, ill and injured employees are forced to stay on the line or be fired, and, yes, doors are still locked from the outside."
An SMT processor can uniquely exploit whichever type of parallelism is available, thereby utilizing the functional units more effectively to achieve the goals of greater throughput and significant program speedups.