Multi-User Computation Partitioning for Latency Sensitive Mobile Cloud Applications

Abstract

Abstract-Elastic partitioning of computations between mobile devices and cloud is an important and challenging research topic for mobile cloud computing. Existing works focus on the single-user computation partitioning, which aims to optimize the application completion time for one particular single user. These works assume that the cloud always has enough resources to execute the computations immediately when they are offloaded to the cloud. However, this assumption does not hold for large scale mobile cloud applications. In these applications, due to the competition for cloud resources among a large number of users, the offloaded computations may be executed with certain scheduling delay on the cloud. Single user partitioning that does not take into account the scheduling delay on the cloud may yield significant performance degradation. In this paper, we study, for the first time, Multi-user Computation Partitioning Problem (MCPP), which considers the partitioning of multiple users' computations together with the scheduling of offloaded computations on the cloud resources. Instead of pursuing the minimum application completion time for every single user, we aim to achieve minimum average completion time for all the users, based on the number of provisioned resources on the cloud. We show that MCPP is different from and more difficult than the classical job scheduling problems. We design an offline heuristic algorithm, namely SearchAdjust, to solve MCPP. We demonstrate through benchmarks that SearchAdjust outperforms both the single user partitioning approaches and classical job scheduling approaches by 10% on average in terms of application delay. Based on SearchAdjust, we also design an online algorithm for MCPP that can be easily deployed in practical systems. We validate the effectiveness of our online algorithm using real world load traces.

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

“…• If a layer is added below the (n 0 + 1)-th layer, condition (31) changes because the amount of data to transmit is reduced 14 . We assume that λ 0 volume of data that originally being processed on the (n 0 + 1)-th layer are offloaded to the added layer.…”

“…This is to say, the right side of (31) becomes larger 14. This is to say, the left side of (31) becomes smaller.…”

“…Therefore, the MEC provides the low-latency and high-efficient data processing due to the proximity of the computing resources [12], [13]. However, most works only consider the MEC servers that directly communicate with the EDs via wireless links to offer the in-proximity services [14]- [16]. Due to the limited computing capacities of these MEC servers, it would be desirable that the data that cannot be processed at the MEC servers can be further offloaded to the upper-layer MEC servers, until to the cloud center (CC).…”

A Generic Framework for Task Offloading in mmWave MEC Backhaul Networks

“…Server Capacity: A BS can process the task if and only if it is co-located with a server as it is realized with constraint (9). Additionally, constraint (10) indicates that a server can process the task subject to its capacity.…”

“…Authors in [5][6][7] studied finite radio-and-computational resource allocation to mobile users in multi-user MEC systems with a single dedicated edge server. MEC server scheduling in multi-user MEC systems has been the topic of research such as [8,9]. Server selection problem in heterogeneous MEC system is the most related research field to our work.…”