TY - GEN
T1 - Empirical analysis of containers on resource constrained IoT gateway
AU - Jeong, Jaeyeop
AU - Raza, Syed Muhammad
AU - Kim, Moonseong
AU - Kang, Byungseok
AU - Jang, Boyun
AU - Choo, Hyunseung
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2020/1
Y1 - 2020/1
N2 - Internet of Things (IoT) for warehouse automation in new generation industries and for home gateway in smart homes have very tight latency requirements for data computation and response. To satisfy the requirements, many studies have moved the computation close to IoT devices by virtualizing the services using containers and deploying them on IoT gateways. They do not account for limited resources at IoT gateways, and without careful provisioning IoT gateways can easily become overloaded and inevitably increase the computation time. Currently there are few studies on performance evaluation of containerized services under constrained resources. To enable efficient service provisioning on IoT gateways, this paper presents the empirical evaluation of Docker Swarm (DS), a container solution, on resource constrained devices such as Raspberry Pi3 boards. We have used multiple open source Intrusion Detection System (IDS) and Deep Learning (DL) based data analytic solutions in our experiments to evaluate creation time, CPU utilization, and memory usage. Our results reveal that creation time and memory usage of service container are critical factors for dynamic provisioning in constrained environments. The lessons learned from our empirical case studies with IDS and DL based data analytic services provide essential guidelines for micro services architectures in IoT gateways.
AB - Internet of Things (IoT) for warehouse automation in new generation industries and for home gateway in smart homes have very tight latency requirements for data computation and response. To satisfy the requirements, many studies have moved the computation close to IoT devices by virtualizing the services using containers and deploying them on IoT gateways. They do not account for limited resources at IoT gateways, and without careful provisioning IoT gateways can easily become overloaded and inevitably increase the computation time. Currently there are few studies on performance evaluation of containerized services under constrained resources. To enable efficient service provisioning on IoT gateways, this paper presents the empirical evaluation of Docker Swarm (DS), a container solution, on resource constrained devices such as Raspberry Pi3 boards. We have used multiple open source Intrusion Detection System (IDS) and Deep Learning (DL) based data analytic solutions in our experiments to evaluate creation time, CPU utilization, and memory usage. Our results reveal that creation time and memory usage of service container are critical factors for dynamic provisioning in constrained environments. The lessons learned from our empirical case studies with IDS and DL based data analytic services provide essential guidelines for micro services architectures in IoT gateways.
KW - Container
KW - Data analytic
KW - Deep learning
KW - Docker swarm
KW - Internet of Things
KW - Intrusion detection system
UR - https://www.scopus.com/pages/publications/85082580266
U2 - 10.1109/ICCE46568.2020.9043056
DO - 10.1109/ICCE46568.2020.9043056
M3 - Conference contribution
AN - SCOPUS:85082580266
T3 - Digest of Technical Papers - IEEE International Conference on Consumer Electronics
BT - 2020 IEEE International Conference on Consumer Electronics, ICCE 2020
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2020 IEEE International Conference on Consumer Electronics, ICCE 2020
Y2 - 4 January 2020 through 6 January 2020
ER -