FLiMS: a Fast Lightweight 2-way Merger for Sorting

Abstract

In this paper, we present FLiMS, a highly-efficient and simple parallel algorithm for merging two sorted lists residing in banked and/or wide memory. On FPGAs, its implementation uses fewer hardware resources than the state-of-the-art alternatives, due to the reduced number of comparators and elimination of redundant logic found on prior attempts. In combination with the distributed nature of the selector stage, a higher performance is achieved for the same amount of parallelism or higher. This is useful in many applications such as in parallel merge trees to achieve high-throughput sorting, where the resource utilisation of the merger is critical for building large trees and internalising the workload for fast computation. Also presented are efficient variations of FLiMS for optimizing throughput for skewed datasets, achieving stable sorting or using fewer dequeue signals. Additionally, FLiMS is shown to perform well as conventional software on modern CPUs supporting single-instruction multiple-data (SIMD) instructions, surpassing the performance of some standard libraries for sorting.

Efficiently Removing Sparsity for High-Throughput Stream Processing

“…To maximise efficiency, parallel algorithms need to match specific platforms. For example, merge trees are still considered in state-of-the-art sorters on FPGAs [4], [5], while on CPUs prominent algorithms include radix sort [6], quicksort [7] and others [8]. Memory use is not the only reason for this discrepancy, but still, the main outcome is that there are different parallel memory access patterns, which can benefit from different memory topologies.…”

FPGA Design for Low Delay Comparison-free, Odd-even Merge Sorter

“…Usui.et.al. (9) proposed a tree-based merge sorter for large-scale sorting and successfully implemented. It was observed that the performance was degraded for the data that are distorted.…”