An FPGA-based change-point detection for 10Gbps packet stream

Takuma Iwata, Kohei Nakamura, Yuta Tokusashi, Hiroki Matsutani

Research output: Contribution to journalArticle

Abstract

In statistical analysis and data mining, change-point detection that identifies the change-points which are times when the probability distribution of time series changes has been used for various purposes, such as anomaly detections on network traffic and transaction data. However, computation cost of a conventional AR (Auto-Regression) model based approach is too high and infeasible for online. In this paper, an AR model based online change-point detection algorithm, called ChangeFinder, is implemented on an FPGA (Field Programmable Gate Array) based NIC (Network Interface Card). The proposed system computes the change-point score from time series data received from 10GbE (10Gbit Ethernet). More specifically, it computes the change-point score at the 10GbE NIC in advance of host applications. It can find change-points on single or multiple streams using a context memory. This paper aims to reduce the host workload and improve change-point detection performance by offloading ChangeFinder algorithm from host to the NIC. As evaluations, change-point detection in the FPGA NIC is compared with a baseline software implementation and those enhanced by two network optimization techniques using DPDK and Netfilter in terms of throughput. The result demonstrates 16.8x improvement in change-point detection throughput compared to the baseline software implementation. It is corresponding to the 10GbE line rate. Performance and area overheads when supporting multiple streams are also evaluated.

Original languageEnglish
Pages (from-to)2366-2376
Number of pages11
JournalIEICE Transactions on Information and Systems
VolumeE102D
Issue number12
DOIs
Publication statusPublished - 2019 Jan 1

Fingerprint

Interfaces (computer)
Field programmable gate arrays (FPGA)
Ethernet
Time series
Throughput
Probability distributions
Data mining
Statistical methods
Data storage equipment
Costs

Keywords

  • 10GbE
  • Change-point detection
  • FPGA NIC

ASJC Scopus subject areas

  • Software
  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering
  • Artificial Intelligence

Cite this

An FPGA-based change-point detection for 10Gbps packet stream. / Iwata, Takuma; Nakamura, Kohei; Tokusashi, Yuta; Matsutani, Hiroki.

In: IEICE Transactions on Information and Systems, Vol. E102D, No. 12, 01.01.2019, p. 2366-2376.

Research output: Contribution to journalArticle

Iwata, Takuma ; Nakamura, Kohei ; Tokusashi, Yuta ; Matsutani, Hiroki. / An FPGA-based change-point detection for 10Gbps packet stream. In: IEICE Transactions on Information and Systems. 2019 ; Vol. E102D, No. 12. pp. 2366-2376.
@article{e9042cb113ef4a68bae77cd8cf5cb5de,
title = "An FPGA-based change-point detection for 10Gbps packet stream",
abstract = "In statistical analysis and data mining, change-point detection that identifies the change-points which are times when the probability distribution of time series changes has been used for various purposes, such as anomaly detections on network traffic and transaction data. However, computation cost of a conventional AR (Auto-Regression) model based approach is too high and infeasible for online. In this paper, an AR model based online change-point detection algorithm, called ChangeFinder, is implemented on an FPGA (Field Programmable Gate Array) based NIC (Network Interface Card). The proposed system computes the change-point score from time series data received from 10GbE (10Gbit Ethernet). More specifically, it computes the change-point score at the 10GbE NIC in advance of host applications. It can find change-points on single or multiple streams using a context memory. This paper aims to reduce the host workload and improve change-point detection performance by offloading ChangeFinder algorithm from host to the NIC. As evaluations, change-point detection in the FPGA NIC is compared with a baseline software implementation and those enhanced by two network optimization techniques using DPDK and Netfilter in terms of throughput. The result demonstrates 16.8x improvement in change-point detection throughput compared to the baseline software implementation. It is corresponding to the 10GbE line rate. Performance and area overheads when supporting multiple streams are also evaluated.",
keywords = "10GbE, Change-point detection, FPGA NIC",
author = "Takuma Iwata and Kohei Nakamura and Yuta Tokusashi and Hiroki Matsutani",
year = "2019",
month = "1",
day = "1",
doi = "10.1587/transinf.2019PAP0015",
language = "English",
volume = "E102D",
pages = "2366--2376",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "12",

}

TY - JOUR

T1 - An FPGA-based change-point detection for 10Gbps packet stream

AU - Iwata, Takuma

AU - Nakamura, Kohei

AU - Tokusashi, Yuta

AU - Matsutani, Hiroki

PY - 2019/1/1

Y1 - 2019/1/1

N2 - In statistical analysis and data mining, change-point detection that identifies the change-points which are times when the probability distribution of time series changes has been used for various purposes, such as anomaly detections on network traffic and transaction data. However, computation cost of a conventional AR (Auto-Regression) model based approach is too high and infeasible for online. In this paper, an AR model based online change-point detection algorithm, called ChangeFinder, is implemented on an FPGA (Field Programmable Gate Array) based NIC (Network Interface Card). The proposed system computes the change-point score from time series data received from 10GbE (10Gbit Ethernet). More specifically, it computes the change-point score at the 10GbE NIC in advance of host applications. It can find change-points on single or multiple streams using a context memory. This paper aims to reduce the host workload and improve change-point detection performance by offloading ChangeFinder algorithm from host to the NIC. As evaluations, change-point detection in the FPGA NIC is compared with a baseline software implementation and those enhanced by two network optimization techniques using DPDK and Netfilter in terms of throughput. The result demonstrates 16.8x improvement in change-point detection throughput compared to the baseline software implementation. It is corresponding to the 10GbE line rate. Performance and area overheads when supporting multiple streams are also evaluated.

AB - In statistical analysis and data mining, change-point detection that identifies the change-points which are times when the probability distribution of time series changes has been used for various purposes, such as anomaly detections on network traffic and transaction data. However, computation cost of a conventional AR (Auto-Regression) model based approach is too high and infeasible for online. In this paper, an AR model based online change-point detection algorithm, called ChangeFinder, is implemented on an FPGA (Field Programmable Gate Array) based NIC (Network Interface Card). The proposed system computes the change-point score from time series data received from 10GbE (10Gbit Ethernet). More specifically, it computes the change-point score at the 10GbE NIC in advance of host applications. It can find change-points on single or multiple streams using a context memory. This paper aims to reduce the host workload and improve change-point detection performance by offloading ChangeFinder algorithm from host to the NIC. As evaluations, change-point detection in the FPGA NIC is compared with a baseline software implementation and those enhanced by two network optimization techniques using DPDK and Netfilter in terms of throughput. The result demonstrates 16.8x improvement in change-point detection throughput compared to the baseline software implementation. It is corresponding to the 10GbE line rate. Performance and area overheads when supporting multiple streams are also evaluated.

KW - 10GbE

KW - Change-point detection

KW - FPGA NIC

UR - http://www.scopus.com/inward/record.url?scp=85076437511&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85076437511&partnerID=8YFLogxK

U2 - 10.1587/transinf.2019PAP0015

DO - 10.1587/transinf.2019PAP0015

M3 - Article

AN - SCOPUS:85076437511

VL - E102D

SP - 2366

EP - 2376

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 12

ER -