Grand challenge: MtDetector: A high-performance marine traic detector at stream scale

Chun Xun Lin, Tsung-Wei Huang, Guannan Guo, Martin D F Wong

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we present MtDetector, a high performance marine trac detector that can predict the destination and the arrival time of travelling vessels. MtDetector accepts streaming data reported by the moving vessels and generates continuous predictions of the arrival port and arrival time for those vessels. To predict the destination for a ship, MtDetector builds a neural network for every port and infers the arrival port for vessels based on their departure port. For the arrival time prediction, we derive informative features from training data and apply Deep Neural Network (DNN) to estimate the traveling time. MtDetector is built on top of DtCraft [1, 2], a high-performance distributed execution engine for stream programming. By utilizing the task-based parallelism in DtCraft, MtDetector can process multiple predictions concurrently to achieve high throughput and low latency.

Original languageEnglish (US)
Title of host publicationDEBS 2018 - Proceedings of the 12th ACM International Conference on Distributed and Event-Based Systems
PublisherAssociation for Computing Machinery, Inc
Pages205-208
Number of pages4
ISBN (Electronic)9781450357821
DOIs
StatePublished - Jun 25 2018
Event12th ACM International Conference on Distributed and Event-Based Systems, DEBS 2018 - Hamilton, New Zealand
Duration: Jun 25 2018Jun 26 2018

Other

Other12th ACM International Conference on Distributed and Event-Based Systems, DEBS 2018
CountryNew Zealand
CityHamilton
Period6/25/186/26/18

Fingerprint

Detectors
Ships
Throughput
Engines
Neural networks
Deep neural networks

Keywords

  • Distributed System
  • Machine Learning
  • Marine Trac
  • Stream Processing

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Computer Networks and Communications
  • Software
  • Hardware and Architecture

Cite this

Lin, C. X., Huang, T-W., Guo, G., & Wong, M. D. F. (2018). Grand challenge: MtDetector: A high-performance marine traic detector at stream scale. In DEBS 2018 - Proceedings of the 12th ACM International Conference on Distributed and Event-Based Systems (pp. 205-208). Association for Computing Machinery, Inc. https://doi.org/10.1145/3210284.3220504

Grand challenge : MtDetector: A high-performance marine traic detector at stream scale. / Lin, Chun Xun; Huang, Tsung-Wei; Guo, Guannan; Wong, Martin D F.

DEBS 2018 - Proceedings of the 12th ACM International Conference on Distributed and Event-Based Systems. Association for Computing Machinery, Inc, 2018. p. 205-208.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Lin, CX, Huang, T-W, Guo, G & Wong, MDF 2018, Grand challenge: MtDetector: A high-performance marine traic detector at stream scale. in DEBS 2018 - Proceedings of the 12th ACM International Conference on Distributed and Event-Based Systems. Association for Computing Machinery, Inc, pp. 205-208, 12th ACM International Conference on Distributed and Event-Based Systems, DEBS 2018, Hamilton, New Zealand, 6/25/18. https://doi.org/10.1145/3210284.3220504
Lin CX, Huang T-W, Guo G, Wong MDF. Grand challenge: MtDetector: A high-performance marine traic detector at stream scale. In DEBS 2018 - Proceedings of the 12th ACM International Conference on Distributed and Event-Based Systems. Association for Computing Machinery, Inc. 2018. p. 205-208 https://doi.org/10.1145/3210284.3220504
Lin, Chun Xun ; Huang, Tsung-Wei ; Guo, Guannan ; Wong, Martin D F. / Grand challenge : MtDetector: A high-performance marine traic detector at stream scale. DEBS 2018 - Proceedings of the 12th ACM International Conference on Distributed and Event-Based Systems. Association for Computing Machinery, Inc, 2018. pp. 205-208
@inproceedings{f3f2a53f393f45f1ab9adb1a714ac978,
title = "Grand challenge: MtDetector: A high-performance marine traic detector at stream scale",
abstract = "In this paper, we present MtDetector, a high performance marine trac detector that can predict the destination and the arrival time of travelling vessels. MtDetector accepts streaming data reported by the moving vessels and generates continuous predictions of the arrival port and arrival time for those vessels. To predict the destination for a ship, MtDetector builds a neural network for every port and infers the arrival port for vessels based on their departure port. For the arrival time prediction, we derive informative features from training data and apply Deep Neural Network (DNN) to estimate the traveling time. MtDetector is built on top of DtCraft [1, 2], a high-performance distributed execution engine for stream programming. By utilizing the task-based parallelism in DtCraft, MtDetector can process multiple predictions concurrently to achieve high throughput and low latency.",
keywords = "Distributed System, Machine Learning, Marine Trac, Stream Processing",
author = "Lin, {Chun Xun} and Tsung-Wei Huang and Guannan Guo and Wong, {Martin D F}",
year = "2018",
month = "6",
day = "25",
doi = "10.1145/3210284.3220504",
language = "English (US)",
pages = "205--208",
booktitle = "DEBS 2018 - Proceedings of the 12th ACM International Conference on Distributed and Event-Based Systems",
publisher = "Association for Computing Machinery, Inc",

}

TY - GEN

T1 - Grand challenge

T2 - MtDetector: A high-performance marine traic detector at stream scale

AU - Lin, Chun Xun

AU - Huang, Tsung-Wei

AU - Guo, Guannan

AU - Wong, Martin D F

PY - 2018/6/25

Y1 - 2018/6/25

N2 - In this paper, we present MtDetector, a high performance marine trac detector that can predict the destination and the arrival time of travelling vessels. MtDetector accepts streaming data reported by the moving vessels and generates continuous predictions of the arrival port and arrival time for those vessels. To predict the destination for a ship, MtDetector builds a neural network for every port and infers the arrival port for vessels based on their departure port. For the arrival time prediction, we derive informative features from training data and apply Deep Neural Network (DNN) to estimate the traveling time. MtDetector is built on top of DtCraft [1, 2], a high-performance distributed execution engine for stream programming. By utilizing the task-based parallelism in DtCraft, MtDetector can process multiple predictions concurrently to achieve high throughput and low latency.

AB - In this paper, we present MtDetector, a high performance marine trac detector that can predict the destination and the arrival time of travelling vessels. MtDetector accepts streaming data reported by the moving vessels and generates continuous predictions of the arrival port and arrival time for those vessels. To predict the destination for a ship, MtDetector builds a neural network for every port and infers the arrival port for vessels based on their departure port. For the arrival time prediction, we derive informative features from training data and apply Deep Neural Network (DNN) to estimate the traveling time. MtDetector is built on top of DtCraft [1, 2], a high-performance distributed execution engine for stream programming. By utilizing the task-based parallelism in DtCraft, MtDetector can process multiple predictions concurrently to achieve high throughput and low latency.

KW - Distributed System

KW - Machine Learning

KW - Marine Trac

KW - Stream Processing

UR - http://www.scopus.com/inward/record.url?scp=85050549983&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85050549983&partnerID=8YFLogxK

U2 - 10.1145/3210284.3220504

DO - 10.1145/3210284.3220504

M3 - Conference contribution

AN - SCOPUS:85050549983

SP - 205

EP - 208

BT - DEBS 2018 - Proceedings of the 12th ACM International Conference on Distributed and Event-Based Systems

PB - Association for Computing Machinery, Inc

ER -