Choice of parallelism: multi-GPU driven pipeline for huge academic backbone network. Issue 6 (2nd November 2021)

• Record Type: Journal Article
• Authors: Ando, Ruo; Kadobayashi, Youki; Takakura, Hiroki
• Is Part Of: International journal of parallel, emergent and distributed systems. Volume 36:Issue 6(2021)
• Page Start: 609
• Page End: 622
• Publication Date: 2021-11-02
• Subjects: Multi-GPU -- streaming data pipeline -- massive session data -- map reduce -- tiling pattern; Parallel computers -- Periodicals; Electronic data processing -- Distributed processing -- Periodicals; Computer algorithms -- Periodicals; 004.35
• DOI: 10.1080/17445760.2021.1941009
• Languages: English
• ISSNs: 1744-5760

Abstract

Abstract : Science Information Network (SINET) is a Japanese academic backbone network for more than 800 research institutions and universities. In this paper, we present a multi-GPU-driven pipeline for handling huge session data of SINET. Our pipeline consists of ELK stack, multi-GPU server, and Splunk. A multi-GPU server is responsible for two procedures: discrimination and histogramming. Discrimination is dividing session data into ingoing/outgoing with subnet mask calculation and network address matching. Histogramming is grouping ingoing/outgoing session data into bins with map-reduce. In our architecture, we use GPU for the acceleration of ingress/egress discrimination of session data. Also, we use a tiling design pattern for building a two-stage map-reduce of CPU and GPU. Our multi-GPU-driven pipeline has succeeded in processing huge workloads of about 1.2–1.6 billion session streams (500–650 GB) within 24 hours. GRAPHICAL ABSTRACT: UF0001