TY - GEN
T1 - M-DB
T2 - 12th IEEE International Conference on Internet of Things, 15th IEEE International Conference on Green Computing and Communications, 12th IEEE International Conference on Cyber, Physical and Social Computing and 5th IEEE International Conference on Smart Data, iThings/GreenCom/CPSCom/SmartData 2019
AU - Arora, Vaibhav
AU - Amiri, Mohammad Javad
AU - Agrawal, Divyakant
AU - El Abbadi, Amr
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/7
Y1 - 2019/7
N2 - IoT devices influence many different spheres of society and are predicted to have a huge impact on our future. Extracting real-Time insights from diverse sensor data and dealing with the underlying uncertainty of sensor data are two main challenges of the IoT ecosystem In this paper, we propose a data processing architecture, M-DB, to effectively integrate and continuously monitor uncertain and diverse IoT data. M-DB constitutes of three components: (1) model-based operators (MBO) as data management abstractions for IoT application developers to integrate data from diverse sensors. Model-based operators can support event-detection and statistical aggregation operators, (2) M-Stream, a dataflow pipeline that combines model-based operators to perform computations reflecting the uncertainty of underlying data, and (3) M-Store, a storage layer separating the computation of application logic from physical sensor data management, to effectively deal with missing or delayed sensor data. M-DB is designed and implemented over Apache Storm and Apache Kafka, two open-source distributed event processing systems. Our illustrated application examples throughout the paper and evaluation results illustrate that M-DB provides a real-Time data-processing architecture that can cater to the diverse needs of IoT applications.
AB - IoT devices influence many different spheres of society and are predicted to have a huge impact on our future. Extracting real-Time insights from diverse sensor data and dealing with the underlying uncertainty of sensor data are two main challenges of the IoT ecosystem In this paper, we propose a data processing architecture, M-DB, to effectively integrate and continuously monitor uncertain and diverse IoT data. M-DB constitutes of three components: (1) model-based operators (MBO) as data management abstractions for IoT application developers to integrate data from diverse sensors. Model-based operators can support event-detection and statistical aggregation operators, (2) M-Stream, a dataflow pipeline that combines model-based operators to perform computations reflecting the uncertainty of underlying data, and (3) M-Store, a storage layer separating the computation of application logic from physical sensor data management, to effectively deal with missing or delayed sensor data. M-DB is designed and implemented over Apache Storm and Apache Kafka, two open-source distributed event processing systems. Our illustrated application examples throughout the paper and evaluation results illustrate that M-DB provides a real-Time data-processing architecture that can cater to the diverse needs of IoT applications.
KW - Abstractions
KW - Iot
KW - Prediction
KW - Real-Time Processing
UR - https://www.scopus.com/pages/publications/85074861973
U2 - 10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00187
DO - 10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00187
M3 - Conference contribution
AN - SCOPUS:85074861973
T3 - Proceedings - 2019 IEEE International Congress on Cybermatics: 12th IEEE International Conference on Internet of Things, 15th IEEE International Conference on Green Computing and Communications, 12th IEEE International Conference on Cyber, Physical and Social Computing and 5th IEEE International Conference on Smart Data, iThings/GreenCom/CPSCom/SmartData 2019
SP - 1096
EP - 1105
BT - Proceedings - 2019 IEEE International Congress on Cybermatics
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 14 July 2019 through 17 July 2019
ER -