International Journal of Parallel Programming · 2022 · 14 citations · 16 references
ComputingComputer EngineeringComputer ScienceCluster ComputingDistributed SystemsParallel ComputingDistributed ComputingDistributed Memory SystemsParallel Computing SystemsParallel Runtime SystemsShared-memory MultiprocessingShared-memory ParallelismDistributed MemoryDistributed Memory CommunicationDistributed-memory ArchitecturesHigh-performance Computing SystemsParallel ProgrammingSerializationShared-memory Architecture
Abstract We present the new distributed-memory run-time system (RTS) of the C++-based open-source structured parallel programming library FastFlow . The new RTS enables the execution of FastFlow shared-memory applications written using its Building Blocks () on distributed systems with minimal changes to the original program. The changes required are all high-level and deal with introducing distributed groups ( dgroup ), i.e., logical partitions of the BBs composing the application streaming graph. A dgroup , which in turn is implemented using FastFlow ’s , can be deployed and executed on a remote machine and communicate with other dgroups according to the original shared-memory FastFlow streaming programming model. We present how to define the distributed groups and how we faced the problem of data serialization and communication performance tuning through transparent messages’ batching and their scheduling. Finally, we present a study of the overhead introduced by dgroups considering some benchmarks on a sixteen-node cluster.
16
Big data analytics on Apache Spark
Salman Salloum, Ruslan Dautov, Xiaojun Chen et al. · International Journal of Data Science and Analytics · 2016 · 425 citations · Full text
Development of Mixed Mode MPI / OpenMP Applications
Lorna Smith, Mark Bull · Scientific Programming · 2001 · 192 citations · Full text