HetMEC: Latency-Optimal Task Assignment and Resource Allocation for Heterogeneous Multi-Layer Mobile Edge Computing

Abstract

Driven by great demands on low-latency services of the edge devices (EDs), mobile edge computing (MEC) has been proposed to enable the computing capacities at the edge of the radio access network. However, conventional MEC servers suffer disadvantages such as limited computing capacity, preventing the computation-intensive tasks to be processed in time. To relief this issue, we propose the heterogeneous MEC (HetMEC) where the data that cannot be timely processed at the edge are allowed be offloaded to the upper-layer MEC servers, and finally to the cloud center (CC) with more powerful computing capacity. We design the latency minimization algorithm by jointly coordinating the task assignment, computing and transmission resources among the EDs, multi-layer MEC servers, and the CC. Simulation results indicate that our proposed algorithm can achieve a lower latency and higher processing rate than the conventional MEC scheme. Index TermsHeterogeneous mobile edge computing, multi-layer MEC, task assignment, resource allocationThe authors are with

Communication-and-Computing Latency Minimization for UAV-Enabled Virtual Reality Delivery Systems

“…The noise spectral density is σ 2 = −169 dBm/Hz. The transmit powers at the U-BS and the cloud server are p u = p c = 0.5 W. We consider the input data size I i follows a uniform distribution with I i ∼ U [10,15] KB, the ratio between O i and I i is set as α = 2, and the required number of CPU cycles per bit is distributed as F i ∼ U [500, 800] cycles/bit. The computing capacity of VR users follows a distribution of f local i ∼ U [0.5, 1] GHz.…”

“…where r i is the transmission rate between the U-BS and the i-th VR user which is shown in (10). The first term in the righthand-side (RHS) of (2) shows that if the data is computed at the U-BS, the output data O i after being processed will be transmitted from U-BS to the VR user and the second term means that if the data is computed locally, the U-BS transmits the input data I i to the VR user for calculation.…”

“…In [9], the authors implemented and developed a low-latency management framework for distributed service function chains enabling tactile internet with MEC. In [10], by jointly coordinating the task assignment, computing, and transmission resources among edge devices, multi-layer MEC servers and cloud center, the authors proposed an efficient algorithm that aimed at minimizing the system latency including total computing and transmission time in heterogeneous multi-layer MEC networks. In [11], by jointly optimizing the users' transmit power, computing capacity allocation, and user association, a latency minimization problem of an MEC system was formulated.…”

Computation offloading through mobile vehicles in IoT-edge-cloud network

“…The IoT-edge-cloud network can make full use of IoT, MEC computing and cloud computing [21][22][23][24][25][26]. Wang et al [27] proposed a task offloading and resource allocation algorithm in a heterogeneous network. The network divided different MEC servers into multiple layers, the objective was to minimize the overall system latency.…”

Computation offloading through mobile vehicles in IoT-edge-cloud network

“…The IoT-edge-cloud network can make full use of WSNs, MEC computing and cloud computing [20][21][22][23][24]. Wang et al [25] proposed a task offloading and resource allocation algorithm in a heterogeneous network. The network divided different MEC servers into multiple layers, the objective were minimize the overall system latency.…”