TY - GEN
T1 - MORA on the Edge
T2 - 11th IEEE International Conference on Cloud Networking, CloudNet 2022
AU - Ouedraogo, Wendlasida
AU - Di Stefano, Alessandro
AU - Araldo, Andrea
AU - Di Stefano, Antonella
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022/1/1
Y1 - 2022/1/1
N2 - In this paper, we present an implementation of the concept of Multiple Option Resource Allocation (MORA) in the context of Multitenant Edge Computing. We assume a Network Operator (NO) owns computational resources on the edge, e.g., storage and CPU co-located with a mobile base station, and opens them to 3rd party Service Providers (SPs), which in this work are video streaming services. Each SP is fully containerized in Kubernetes pods. Following MORA concept, in order to cope with the limited resources on the edge, we assume that each SP can run under different configurations, consuming more or less amount of resources, so as to adapt to the actual resource availability. Therefore, the NO can choose the best configuration per each SP, in order to optimize some NO's related performance (the upstream traffic, in our case).We implement a test-bed showcasing such a system and we measure its performance. All the code is released as open source [1].
AB - In this paper, we present an implementation of the concept of Multiple Option Resource Allocation (MORA) in the context of Multitenant Edge Computing. We assume a Network Operator (NO) owns computational resources on the edge, e.g., storage and CPU co-located with a mobile base station, and opens them to 3rd party Service Providers (SPs), which in this work are video streaming services. Each SP is fully containerized in Kubernetes pods. Following MORA concept, in order to cope with the limited resources on the edge, we assume that each SP can run under different configurations, consuming more or less amount of resources, so as to adapt to the actual resource availability. Therefore, the NO can choose the best configuration per each SP, in order to optimize some NO's related performance (the upstream traffic, in our case).We implement a test-bed showcasing such a system and we measure its performance. All the code is released as open source [1].
KW - Content Delivery
KW - Edge computing
KW - Multitenant
U2 - 10.1109/CloudNet55617.2022.9978894
DO - 10.1109/CloudNet55617.2022.9978894
M3 - Conference contribution
AN - SCOPUS:85146121548
T3 - Proceedings of the 2022 IEEE Conference on Cloud Networking 2022, CloudNet 2022
SP - 262
EP - 266
BT - Proceedings of the 2022 IEEE Conference on Cloud Networking 2022, CloudNet 2022
A2 - Secci, Stefano
A2 - Durairajan, Ramakrishnan
A2 - Linguaglossa, Leonardo
A2 - Kamiyama, Noriaki
A2 - Nogueira, Michele
A2 - Rovedakis, Stephane
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 7 November 2022 through 10 November 2022
ER -