An FPGA-Based Change-Point Detection for 10Gbps Packet Stream

Takuma IWATA; Kohei NAKAMURA; Yuta TOKUSASHI; Hiroki MATSUTANI

doi:10.1587/transinf.2019PAP0015

IEICE TRANSACTIONS on Information

Open Access
An FPGA-Based Change-Point Detection for 10Gbps Packet Stream

Takuma IWATA, Kohei NAKAMURA, Yuta TOKUSASHI, Hiroki MATSUTANI

Full Text Views

25

Cite this

Free PDF (1.7MB)

Summary :

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.

Publication: IEICE TRANSACTIONS on Information Vol.E102-D No.12 pp.2366-2376

Publication Date: 2019/12/01

Publicized: 2019/07/23

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2019PAP0015

Type of Manuscript: Special Section PAPER (Special Section on Parallel and Distributed Computing and Networking)

Category: Computer System

Authors

Takuma IWATA
  Keio University
Kohei NAKAMURA
  Keio University
Yuta TOKUSASHI
  Keio University
Hiroki MATSUTANI
  Keio University

Keyword

Change-point detection, FPGA NIC, 10GbE

Cite this

Copy

Takuma IWATA, Kohei NAKAMURA, Yuta TOKUSASHI, Hiroki MATSUTANI, "An FPGA-Based Change-Point Detection for 10Gbps Packet Stream" in IEICE TRANSACTIONS on Information, vol. E102-D, no. 12, pp. 2366-2376, December 2019, doi: 10.1587/transinf.2019PAP0015.
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.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2019PAP0015/_p

Copy

@ARTICLE{e102-d_12_2366,
author={Takuma IWATA, Kohei NAKAMURA, Yuta TOKUSASHI, Hiroki MATSUTANI, },
journal={IEICE TRANSACTIONS on Information},
title={An FPGA-Based Change-Point Detection for 10Gbps Packet Stream},
year={2019},
volume={E102-D},
number={12},
pages={2366-2376},
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={},
doi={10.1587/transinf.2019PAP0015},
ISSN={1745-1361},
month={December},}

Copy

TY - JOUR
TI - An FPGA-Based Change-Point Detection for 10Gbps Packet Stream
T2 - IEICE TRANSACTIONS on Information
SP - 2366
EP - 2376
AU - Takuma IWATA
AU - Kohei NAKAMURA
AU - Yuta TOKUSASHI
AU - Hiroki MATSUTANI
PY - 2019
DO - 10.1587/transinf.2019PAP0015
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E102-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2019
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.
ER -

IEICE TRANSACTIONS on Information