TY - JOUR
T1 - Optimized provisioning of edge computing resources with heterogeneous workload in IoT networks
AU - Kherraf, Nouha
AU - Alameddine, Hyame Assem
AU - Sharafeddine, Sanaa
AU - Assi, Chadi M.
AU - Ghrayeb, Ali
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/6
Y1 - 2019/6
N2 - The proliferation of smart connected Internet of Things (IoT) devices is bringing tremendous challenges in meeting the performance requirement of their supported real-time applications due to their limited resources in terms of computing, storage, and battery life. In addition, the considerable amount of data they generate brings extra burden to the existing wireless network infrastructure. By enabling distributed computing and storage capabilities at the edge of the network, multi-access edge computing (MEC) serves delay sensitive, computationally intensive applications. Managing the heterogeneity of the workload generated by IoT devices, especially in terms of computing and delay requirements, while being cognizant of the cost to network operators, requires an efficient dimensioning of the MEC-enabled network infrastructure. Hence, in this paper, we study and formulate the problem of MEC resource provisioning and workload assignment for IoT services (RPWA) as a mixed integer program to jointly decide on the number and the location of edge servers and applications to deploy, in addition to the workload assignment. Given its complexity, we propose a decomposition approach to solve it which consists of decomposing RPWA into the delay aware load assignment sub-problem and the mobile edge servers dimensioning sub-problem. We analyze the effectiveness of the proposed algorithm through extensive simulations and highlight valuable performance trends and trade-offs as a function of various system parameters.
AB - The proliferation of smart connected Internet of Things (IoT) devices is bringing tremendous challenges in meeting the performance requirement of their supported real-time applications due to their limited resources in terms of computing, storage, and battery life. In addition, the considerable amount of data they generate brings extra burden to the existing wireless network infrastructure. By enabling distributed computing and storage capabilities at the edge of the network, multi-access edge computing (MEC) serves delay sensitive, computationally intensive applications. Managing the heterogeneity of the workload generated by IoT devices, especially in terms of computing and delay requirements, while being cognizant of the cost to network operators, requires an efficient dimensioning of the MEC-enabled network infrastructure. Hence, in this paper, we study and formulate the problem of MEC resource provisioning and workload assignment for IoT services (RPWA) as a mixed integer program to jointly decide on the number and the location of edge servers and applications to deploy, in addition to the workload assignment. Given its complexity, we propose a decomposition approach to solve it which consists of decomposing RPWA into the delay aware load assignment sub-problem and the mobile edge servers dimensioning sub-problem. We analyze the effectiveness of the proposed algorithm through extensive simulations and highlight valuable performance trends and trade-offs as a function of various system parameters.
KW - 5G
KW - Internet of Things
KW - Multi-access edge computing
KW - Operation research
KW - Optimization
KW - Resource allocation
KW - Task offloading
UR - http://www.scopus.com/inward/record.url?scp=85071107031&partnerID=8YFLogxK
U2 - 10.1109/TNSM.2019.2894955
DO - 10.1109/TNSM.2019.2894955
M3 - Article
AN - SCOPUS:85071107031
SN - 1932-4537
VL - 16
SP - 459
EP - 474
JO - IEEE Transactions on Network and Service Management
JF - IEEE Transactions on Network and Service Management
IS - 2
M1 - 2894955
ER -