Discrete-event simulation for efficient and stable resource allocation in collaborative mobile cloudlets

Abstract

The deployment of highly interactive, media-rich applications on mobile devices is hindered by the inherent limitations on compute power, memory and battery capacity of these hand-held platforms. The cloudlet concept, opportunistically offloading computation to nearby devices, has proven to be a viable solution in offering resourceintensive applications on mobile devices. In this paper, we propose to extend the cloudlet concept with collaborative scenarios, in which not only hardware resources for processing are shared between all cloudlet users, but also the data computed.In a cloudlet, the resource demand should be spread over all available cloudlet nodes. User mobility and fluctuations in wireless bandwidth will cause the optimal resource allocation to vary over time. The cloudlet middleware must continuously balance the performance gain of reallocating components with the operational costs in terms of user experience and management complexity. In this paper, we formulate this optimization problem based on a theoretical cloudlet model capturing the infrastructure, application structure and user behaviour.In order to solve this problem, two heuristic allocation algorithms based on Steepest Descent (SD) and Simulated Annealing (SA) are described. Besides optimality of the found solution, it is also important to limit the number of reallocations at runtime. To evaluate the performance and stability of the algorithms, we propose a discreteevent model for cloudlet simulation. For multiple application scenarios, we observe that SD performs 4 times less reallocations than SA. By introducing hysteresis, the number of reallocations by SA can be nearly halved without any significant degradation of application performance.

Dynamic energy-aware cloudlet-based mobile cloud computing model for green computing

“…This approach was efficient in a specific condition that implies the users can accept a scalable service delivery manner during the usages. Some other research had other focuses, such as using cloudlets for efficient data allocation (Quwaider and Jararweh, 2015), resource allocation (Bohez et al, 2015), or admission control policy (Hoang et al, 2012). Nonetheless, none of them had the same target as ours.…”

Decentralized Algorithm for Randomized Task Allocation in Fog Computing Systems

“…Furthermore, assuming that the first moment X i and the second moment 2 X i of X i can be estimated based on the past, the statistics of the number of CPU cycles required to perform the task of device i can be easily obtained. Similar to other works [17], [18], [19], we assume that the task arrival process of device i can be modeled by a Poisson process with arrival intensity λ i . For each task t i,k device i can decide whether to perform the task locally, to offload it to a device j ∈ N \ {i} or to a cloud server.…”

Optimized Cloudlet Scheduling Based Method for Mobile Phones

“…The cloudlet platforms provide relatively moderate power consumption compared to the cloning technology. The existence of steady bandwidth and latency link for the uplink and downlink associated with task processing dictates the magnitude of energy and time spent by the signal in accomplishing the task designated in the architecture [2].…”

“…The algorithm facilitates the decision-making process due to the increased use of relatively efficient means that foster the development consistency processes without ignoring the problem statements suggested in the solution section. The algorithm compares the intended outcome to the general objectives based on the potential improvement in data security and service delivery [2]. The tests conducted by the algorithm during the development stage, eliminate the weaknesses and error incurred during the formulation of algorithm confine the task in a particular process that improves the analysis, control, and evaluation of the codes before implementation in the communication platform.…”