Parallel mining of time-faded heavy hitters
作者：
Highlights：
• We introduce PFDCMSS, a parallel algorithm for mining time-faded heavy hitters.
• The correctness of the algorithm is formally proved.
• The mergeability of augmented sketch data structures is non trivial.
• Experimental results demonstrate that PFDCMSS provides extreme accuracy.
• The message-passing based parallel approach provides excellent scalability.
摘要
•We introduce PFDCMSS, a parallel algorithm for mining time-faded heavy hitters.•The correctness of the algorithm is formally proved.•The mergeability of augmented sketch data structures is non trivial.•Experimental results demonstrate that PFDCMSS provides extreme accuracy.•The message-passing based parallel approach provides excellent scalability.
论文关键词：Message–passing,Heavy hitters,Time–fading model,Sketches
论文评审过程：Received 16 January 2017, Revised 9 November 2017, Accepted 10 November 2017, Available online 14 November 2017, Version of Record 22 December 2017.
论文官网地址：https://doi.org/10.1016/j.eswa.2017.11.021
