On superlinear speedups

DOI10.1016/S0167-8191(05)80060-2MaRDI QIDQ1179219zbMATH OpenOpenAlexFDO

Authors Dietrich Fischer

Publication date 26 June 1992

Published in Parallel Computing (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/s0167-8191(05)80060-2

zbMATH Keywords

efficiencyparallel processorssuperlinear speedup

Mathematics Subject Classification ID

Mathematical problems of computer architecture (68M07)

Review text

\textit{R. Janssen} [Parallel Comput. 4, 211-213 (1987; Zbl 0641.68070)] proposed an algorithm for which the use of \(n\) processors can reduce computer time by more than a factor of \(n\). It is shown that in the example that Janssen offered the superlinear speedup was possible only due to inefficient programming of the algorithm on a single processor. Yet Janssen's claim is correct, as shown by another example.

Recommendations

• A note on superlinear speedup
• Superlinear performance in real-time parallel computation
• scientific article; zbMATH DE number 4178775
• scientific article; zbMATH DE number 4090815
• scientific article; zbMATH DE number 4096771

Cited in

(9)

• A note on superlinear speedup
• Speedup and efficiency of large-size applications on heterogeneous networks
• About \textsc{Fast} and \textsc{TReX} accelerations
• scientific article; zbMATH DE number 4178775 (Why is no real title available?)
• scientific article; zbMATH DE number 4096771 (Why is no real title available?)
• Speedup for natural problems and noncomputability
• A speed-up theorem without tape compression
• Superlinear performance in real-time parallel computation
• Superefficiency from the vantage point of computability

This page was built for publication: On superlinear speedups

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1179219)