Optimization of data exchange in parallel computers with distributed memory

DOI10.1007/S10559-006-0066-7MaRDI QIDQ2371693zbMATH OpenOpenAlexFDO

Authors E. V. Adutskevich, N. A. Likhoded

Publication date 5 July 2007

Published in Cybernetics and Systems Analysis (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/s10559-006-0066-7

zbMATH Keywords

parallel computationsoptimization of communicationsparallel computer with distributed memoryparallelization of algorithmsaffine cycle nests

Mathematics Subject Classification ID

Modes of computation (nondeterministic, parallel, interactive, probabilistic, etc.) (68Q10) Parallel algorithms in computer science (68W10) Distributed systems (68M14)

Recommendations

• Design and optimization of parallel computers for processing mass data
• scientific article; zbMATH DE number 4058853
• Multi-dimensional parallelization considering conditions of data localization
• Parallelization of sequential programs: distribution of arrays among processors and structurization of communications
• Conditions for the localization of data in sequential and parallel implementation of algorithms

Cites work

• Compiling affine nested loops: How to optimize the residual communications after the alignment phase
• Distribution of operations and data arrays over processors
• Finding and using directed cuts of real graphs of algorithms
• Mapping affine loop nests
• Mapping of affine loop nests onto independent processors
• Resource-constrained scheduling of partitioned algorithms on processor arrays

Cited in

(16)

• Minimization of memory and network contention for accessing arbitrary data patterns in SIMD systems
• scientific article; zbMATH DE number 1919438 (Why is no real title available?)
• Optimization of parallel common memory interface for multiprocessor computer systems
• Memory management algorithms for optimistic parallel simulation
• Conditions for the localization of data in sequential and parallel implementation of algorithms
• Sufficient conditions for the determination and use of data in the same granular parallel computation process
• Optimizing large data transfers in parity-declustered data layouts
• Parallelization of sequential programs: distribution of arrays among processors and structurization of communications
• Deriving array distributions by optimization techniques
• Towards Practical Feasibility of Core Computation in Data Exchange
• Optimizing the use of shared data by interacting parallel processes
• Multi-dimensional parallelization considering conditions of data localization
• Automation of the Application of Data Distribution with Overlapping in Distributed Memory
• Design and optimization of parallel computers for processing mass data
• Conditions for the existence of broadcast and spatial locality in computation threads
• Operator composition in a multidimensional structurally oriented model of parallel computations and systems

This page was built for publication: Optimization of data exchange in parallel computers with distributed memory

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2371693)