Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
FlowPools, proposed by are a powerful way to express dataflow logic in highly parallelized applications. The original paper proposes two ways of implementing a FlowPool: Single-Lane FlowPools (SLFP) and Multi-Lane FlowPools (MLFP). While SLFPs showed decent performance overall, insertion operations do not scale. MLFPs solve this limitation as benchmarks discussed in have shown. This report goes into the details of the implementation of MLFPs and lies out the benchmarking results to show that MLFPs may reduce insertion time by 49 - 54% on a 4-core i7 machine with respect to comparable concurrent queue data structures in the Java standard library.
Aurélien François Gilbert Bloch