Offloading Solution Research Articles

Evolving the Cloud Block Store with Performance, Elasticity, Availability, and Hardware Offloading

In this paper, we qualitatively and quantitatively discuss the design choices, production experience, and lessons in building the Elastic Block Storage ( EBS ) at Alibaba Cloud over the past decade. To cope with hardware advancement and users’ demands, we shift our focus from design simplicity in EBS1 to high performance and space efficiency in EBS2 , and finally reducing network traffic amplification in EBS3 . In addition to the architectural evolutions, we also summarize development lessons and experiences as four topics, including: (i) achieving high elasticity in latency, throughput, IOPS and capacity; (ii) improving availability by minimizing the blast radius of individual, regional, and global failure events; (iii) identifying the motivations and key tradeoffs in various hardware offloading solutions; and (iv) identifying the pros/cons of alternative solutions and explaining why seemingly promising ideas would not work in practice.

Delay-aware resource allocation for partial computation offloading in mobile edge cloud computing

Mobile Edge Cloud Computing (MECC), as a promising partial computing offloading solution, has provided new possibilities for compute-intensive and delay-sensitive mobile applications, which can simultaneously leverage edge computing and cloud services. However, designing resource allocation strategies for MECC faces an extremely challenging problem of simultaneously satisfying the end-to-end latency requirements and minimum resource allocation of multiple mobile applications. To address this issue, we comprehensively consider the randomness of computing request arrivals, service time, and dynamic computing resources. We model the MECC network as a two-level tandem queue consisting of two sequential computing processing queues, each with multiple servers. We apply a deep reinforcement learning algorithm called Deep Deterministic Policy Gradient (DDPG) to learn the computing speed adjustment strategy for the tandem queue. This strategy ensures the end-to-end latency requirements of multiple mobile applications while preventing overuse of the total computing resources of edge servers and cloud servers. Numerous simulation experiments demonstrate that our approach is significantly superior to other methods in dynamic network environments.

ASAP: IEEE 802.11ax-based seamless access point handover for moving vehicles

The increasing number of connected and automated vehicles has led to a sharp increase in the demand for network access of moving vehicles. Although 5G networks support terminals with high mobility, the traffic load is too heavy to bear if all the vehicles have a large amount of data for transmission. Therefore, IEEE 802.11-based wireless network is a complementary offload solution to provide high-speed network access for vehicles with low cost, easy deployment and high scalability. However, frequent network handover of moving vehicles between multiple roadside access points (APs) results in network performance degradation, which is one of the challenges in vehicular communications. In this paper, we propose a framework (referred to as ASAP) based on the up-to-date IEEE 802.11ax standard to provide moving vehicles with seamless handover between multiple APs. By leveraging the high efficiency (HE) sounding protocol of IEEE 802.11ax, each AP is capable to monitor the current location of moving vehicles in real time. In addition, a mechanism is also proposed for AP uplink/downlink transmissions through collaboration between the APs and the backbone network to achieve seamless handover for moving vehicles. Since ASAP is based on IEEE 802.11ax, the compatible security scheme such as IEEE 802.11i can be applied to ASAP for security enhancement. The proposed solution does not require any modification on the user terminals, making it possible to be implemented in practice. Extensive simulations show that ASAP significantly reduces the network handover delay to microsecond level, and improves network throughput up to 59% compared with the state-of-the-art methods.

HybOff: a Hybrid Offloading approach to improve load balancing in fog environments

Load balancing is crucial in distributed systems like fog computing, where efficiency is paramount. Offloading with different approaches is the key to balancing the load in distributed environments. Static offloading (SoA) falls short in heterogeneous networks, necessitating dynamic offloading to reduce latency in time-sensitive tasks. However, prevalent dynamic offloading (PoA) solutions often come with hidden costs that impact sensitive applications, including decision time, networks congested and distance offloading. This paper introduces the Hybrid Offloading (HybOff) algorithm, which substantially enhances load balancing and resource utilization in fog networks, addressing issues in both static and dynamic approaches while leveraging clustering theory. Its goal is to create an uncomplicated low-cost offloading approach that enhances IoT application performance by eliminating the consequences of hidden costs regardless of network size. Experimental results using the iFogSim simulation tool show that HybOff significantly reduces offloading messages, distance, and decision-offloading consequences. It improves load balancing by 97%, surpassing SoA (64%) and PoA (88%). Additionally, it increases system utilization by an average of 50% and enhances system performance 1.6 times and 1.4 times more than SoA and PoA, respectively. In summary, this paper tries to introduce a new offloading approach in load balancing research in fog environments.

Simulation Analysis and Comparative Study of MOSGT with Other Techniques

This paper presents a comparative study of the Multi-Objective Symmetric Game Theory (MOSGT) for task offloading in cloud robots with existing techniques such as Genetic Algorithm 4 Collaborative Computation Offloading (GA4CCO) and Approximation Collaborative Computation Offloading (ACCO). The study employs a simulation analysis to evaluate the performance of the proposed MOSGT model in a cloud robotics context. The results of the simulation study reveal that the MOSGT model reduces system cost by approximately 35% compared to GA4CCO and by 20% compared to ACCO. Furthermore, the MOSGT model exhibits a reduced rate of 30% less than GA4CCO and 25% minimal than ACCO in terms of iteration. These findings underscore the potential of MOSGT as a promising solution for efficient task offloading in cloud robots, offering significant improvements in computational complexity and service quality in edge computing.

Task offloading scheme of vehicular cloud edge computing based on Digital Twin and improved A3C

In traditional vehicle edge computing research, the limited computing resources of edge servers are not fully considered. At the same time, when using deep reinforcement learning (DRL) to solve vehicle edge computing problems, it is not fully considered that DRL requires a large amount of real-time data and is prone to gradient explosion or disappearance, over-fitting and local optimal solution problems. Therefore, this article proposes a digital twin (DT)-assisted vehicular cloud edge computing offloading solution. To solve the problem of limited edge server computing resources, this article proposes a vehicle, cloud, and edge server collaboration solution. In response to the problem that DRL requires a large amount of real-time data, this paper proposes a real-time data acquisition method based on DT technology. Aiming at the gradient explosion or disappearance, overfitting, and local optimal solution problems of the asynchronous dominant actor–critic (A3C) algorithm, this paper proposes an improved algorithm of A3C, which introduces the ɛ-greedy strategy and dynamic baseline to increase the possibility of exploration and increase the generalization ability of the model through Dropout technology. During the training process, this article also uses the stochastic gradient descent (SGD) algorithm to accelerate the training process and reduce the computational complexity. The simulation results show that compared with other solutions, the improved algorithm proposed in this paper can effectively reduce the delay and energy consumption of offloading tasks at the vehicular edge cloud.

Deep Meta Q-Learning Based Multi-Task Offloading in Edge-Cloud Systems

Resource-constrained edge devices can not efficiently handle the explosive growth of mobile data and the increasing computational demand of modern-day user applications. Task offloading allows the migration of complex tasks from user devices to the remote edge-cloud servers thereby reducing their computational burden and energy consumption while also improving the efficiency of task processing. However, obtaining the optimal offloading strategy in a multi-task offloading decision-making process is an NP-hard problem. Existing Deep learning techniques with slow learning rates and weak adaptability are not suitable for dynamic multi-user scenarios. In this article, we propose a novel deep meta-reinforcement learning-based approach to the multi-task offloading problem using a combination of first-order meta-learning and deep Q-learning methods. We establish the meta-generalization bounds for the proposed algorithm and demonstrate that it can reduce the time and energy consumption of IoT applications by up to 15%. Through rigorous simulations, we show that our method achieves near-optimal offloading solutions while also being able to adapt to dynamic edge-cloud environments.

Toward Reliable DNN-Based Task Partitioning and Offloading in Vehicular Edge Computing

Modern vehicles have become typical consumer electronics with the development of sensing, transmission, and computation technologies. The emerging intelligent transportation systems (ITSs) yield lots of deep neural network (DNN) based tasks, requiring intensive computation. In view of this, this paper makes the first exploration of accelerating the processing of DNN-based tasks while maintaining decent system reliability in vehicular edge computing (VEC) via task partitioning and offloading. Specifically, we present a specific scenario, where vehicles partition and offload their tasks to nearby vehicles and roadside infrastructures, while they may fail to receive results due to unstable vehicular communications. Then, we model the task delay by considering task properties and node capacities. On this basis, the Partitioning and Offloading Problem (POP) is formulated as a bi-objective optimization problem, to maximize both the acceleration ratio and service ratio of DNN-based tasks in VEC. Further, we propose a Distributed Partitioning and Offloading Solution (DPOS), where a delay-priority-oriented offloading strategy is designed to help edge nodes make offloading decisions, and a stacking-based partitioning strategy is designed to assist client vehicles to make partitioning decisions. Finally, we give a comprehensive performance evaluation, which demonstrates the superiority of the proposed solution.

Multi-RAT-enabled edge computing for vehicle-to-everything architectures

The accelerating deployment of vehicular networks in smart cities sparked a demand for a wider diversity of on-board applications, extending their performance requirements. Higher computation needs and stricter delay constraints provide new challenges for edge computing vehicular architectures, especially regarding urgent data and high-priority tasks. In that regard, QoS-provisioning appears as imperative, along with optimized resource requesting at RSUs. To this end, we propose CAVTOMEC, a multi-RAT location-aware, context-aware task offloading solution with QoS provisioning for MEC vehicular networks. Three concurrent mechanisms are at play: traffic classification, CAM-beacon-enhanced location awareness and long-range V2N resource polling. Three-tier traffic classification identifies task priority, while location and resource awareness participate in the selection of the most appropriate offloading destination depending on these priorities. The resource awareness mechanism is developed in two phases: immediate available resources and prospective resources based on edge server queue state. Experimental results corroborate the gains of our scheme compared to a standard offloading scheme, especially for high-priority tasks, with success rates increased by up to 14%, and offloading delays reduced by up to 24%.

A Markov Decision Process Solution for Energy-Saving Network Selection and Computation Offloading in Vehicular Networks

Vehicular Edge Computing (VEC) enables the integration of edge computing facilities in vehicular networks (VNs), allowing data-intensive and latency-critical applications and services to end-users. Though VEC brings several benefits in terms of reduced task computation time, energy consumption, backhaul link congestion, and data security risks, VEC servers are often resource-constrained. Therefore, the selection of proper edge nodes and the amount of data to be offloaded becomes important for having VEC process benefits. However, with the involvement of high mobility vehicles and dynamically changing vehicular environments, proper VEC node selection and data offloading can be challenging. In this work, we consider a joint network selection and computation offloading problem over a VEC environment for minimizing the overall latency and energy consumption during vehicular task processing, considering both user and infrastructure side energy-saving mechanisms. We have modeled the problem as a sequential decision-making problem and incorporated it in a Markov Decision Process (MDP). Numerous vehicular scenarios are considered based upon the users' positions, the states of the surrounding environment, and the available resources for creating a better environment model for the MDP analysis. We use a value iteration algorithm for finding an optimal policy of the MDPs over an uncertain vehicular environment. Simulation results show that the proposed approaches improve the network performance in terms of latency and consumed energy.

QoS-SLA-aware adaptive genetic algorithm for multi-request offloading in integrated edge-cloud computing in Internet of vehicles

The Internet of Vehicles over vehicular ad hoc network is an emerging technology enabling the development of smart applications focused on improving traffic safety, traffic efficiency, and the overall driving experience. These applications have stringent requirements detailed in the Service Level Agreement. Since vehicles have limited computational and storage capabilities, applications' requests are offloaded onto an integrated edge-cloud computing system. Existing offloading solutions focus on optimizing the application's Quality of Service (QoS) in terms of execution time while respecting a single SLA constraint. They do not consider the impact of overlapped multi-request processing nor the vehicle's varying speed. This paper proposes a novel Artificial Intelligence QoS-SLA-aware adaptive genetic algorithm (QoS-SLA-AGA) to optimize the application's execution time for multi-request offloading in a heterogeneous edge-cloud computing system, which considers the impact of processing multi-requests overlapping and dynamic vehicle speed. The proposed genetic algorithm integrates an adaptive penalty function to assimilate the SLA constraints regarding latency, processing time, deadline, CPU, and memory requirements. Numerical experiments and analysis compare our QoS-SLA-AGA to baseline genetic-based, meta-heuristic Particle Swarm Optimization (PSO), random offloading, All Edge Computing (AEC), and All Cloud Computing (ACC) approaches. Results show QoS-SLA-AGA executes the requests 1.04, 1.23, 1.05, and 9.41 times faster on average compared to the PSO, random offloading, ACC, and AEC approaches respectively. Moreover, the proposed algorithm violates 49.58%, 60.36%, 16.26%, and 80.42% fewer SLAs compared to PSO, random, ACC, and AEC respectively. In contrast, the baseline genetic-based approach increases the requests' performance by 1.14 times, with 24.03% more SLA violations.

Availability-aware multi-component V2X application placement

Vehicle-to-everything (V2X) applications are an integral part of the realization of autonomous and connected vehicles. Due to the large amount of data generated by the vehicles, the stringent delay requirements, and the sheer size of V2X applications, these applications are decomposed into several components that need to be placed in close proximity to requesting vehicles. Mobile-edge Computing (MEC) has been envisioned as a promising solution for vehicular task offloading. Despite the desirable properties of MEC, which can potentially satisfy the delay requirements, edge servers have limited computational capability and have a high cost. They also have availability concerns that should be taken into account. In this regard, Application Service Providers (ASPs) strive for a solution that can both satisfy the delay requirements of V2X applications and cut their expenses, which is reflected by the utilization of edge resources. To that end, this work formulates a Binary Linear Program (BLP) that minimizes the ratio of the experienced to the tolerable delay threshold and the utilization of edge resources under resource and V2X applications' availability constraints. To investigate the trade-off between these two objectives, they are parametrized by a coefficient that reflects their relative priorities. Additionally, the delay and the resource objectives are studied in light of different V2X applications' availability constraints. The experimental results show the effectiveness of our approach. On average, the delays of all V2X applications are satisfied with an acceptable level of edge server utilization.

Decentralized Scheduling and Dynamic Pricing for Edge Computing: A Mean Field Game Approach

Edge computing provides a platform facilitating edge servers to contribute to computation offloading while economizing their resources. Traditional offloading solutions are mostly centralized, which are unscalable for large-scale edge computing networks due to complex interactions among many edge servers. Meanwhile, dynamic pricing for an operator is equally, if not more, important to accommodate users’ time-varying demands for computing services. In this paper, we develop a decentralized online optimization framework to jointly minimize the server’s cost of workload scheduling while maximizing the operator’s utility of service pricing. Specifically, we employ the mean field game to model the collective scheduling behavior of all edge servers, thereby enabling optimal decision making only based on the server’s local information. Considering the service price in practice is not adjusted as frequently as the scheduling process, we establish a two-timescale optimization framework, where workload scheduling at a small timescale is tightly embedded into service pricing at a large timescale. Using mean field approximation, we derive the closed-form expression for the minimum scheduling cost, and the approximation error is <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$O\left(\frac{1}{\sqrt{M}}\right)$</tex-math> </inline-formula> which declines as the number of edge servers <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$M$</tex-math> </inline-formula> increases. By characterizing the influence of workload scheduling on dynamic pricing, we transform the complex service utility maximization into a succinct but equivalent problem, and thus we can make use of Lyapunov optimization to determine the optimal price over time. Extensive evaluations validate the effectiveness and optimality of our scheduling and pricing schemes.

Large-Scale Computation Offloading Using a Multi-Agent Reinforcement Learning in Heterogeneous Multi-Access Edge Computing

Recently, existing computation offloading methods have provided extremely low service latency for mobile users (MUs) in multi-access edge computing (MEC). However, this remains a challenge in large-scale mixed cooperative-competitive MUs heterogeneous MEC environments. Moreover, existing methods focus more on all offloaded tasks handled by static resource allocation MEC servers (ESs) within a time interval, ignoring on-demand requirements of heterogeneous tasks, resulting in many tasks being dropped or wasting resources, especially for latency-sensitive tasks. To address these issues, we present a decentralized computation offloading solution based on the Attention-weighted Recurrent Multi-Agent Actor-Critic (ARMAAC). First, we design a recurrent actor-critic framework to assist MU agents in remembering historical resource allocation information of ESs to better understand the future state of ESs, especially in dynamic resource allocation. Second, an attention mechanism is introduced to compress the joint observation space dimension of all MUs agent to adapt to large-scale MUs. Finally, the actor-critic framework with double centralized critics and Dueling network is redesigned considering the instability and convergence difficulties caused by the sensitive relationship between the actor and critic networks. The experiments show that ARMAAC improves task completion rates and reduces average system cost by 11.01%~14.03% and 10.45%~15.56% compared with baselines.

Dynamic collaborative task offloading for delay minimization in the heterogeneous fog computing systems

Fog computing systems have been widely integrated in IoT-based applications to improve quality of services (QoS) such as low response service delays. This improvement is enabled by task offloading schemes, which perform task computation near the task generation sources (i.e., IoT devices) on behalf of remote cloud servers. However, reducing delay remains challenging for offloading strategies owing to the resource limitations of fog devices. In addition, a high rate of task requests combined with heavy tasks (i.e., large task size) may cause a high imbalance of the workload distribution among the heterogeneous fog devices, which severely impacts the offloading performance in terms of delay. To address these issues, this paper proposes a dynamic collaborative task offloading (DCTO) approach, which is based on the resource states of fog devices, to dynamically derive the task offloading policy. Accordingly, a task can be executed by either a single fog or multiple fog devices through the parallel computation of subtasks to reduce the task execution delay. Through extensive simulation analysis, the proposed offloading solution showed potential advantages in reducing the average delay significantly in systems with a high rate of service requests and heterogeneous fog environment compared with the existing solutions. In addition, the proposed scheme can be implemented online owing to its low computational complexity compared with the algorithms proposed in related works.

Nimbus: Towards Latency-Energy Efficient Task Offloading for AR Services

Widespread adoption of mobile augmented reality (AR) and virtual reality (VR) applications depends on their smoothness and immersiveness. Modern AR applications applying computationally intensive computer vision algorithms can burden today's mobile devices, and cause high energy consumption and/or poor performance. To tackle this challenge, it is possible to offload part of the computation to nearby devices at the edge. However, this calls for smart task placement strategies in order to efficiently use the resources of the edge infrastructure. In this paper, we introduce Nimbus --- a task placement and offloading solution for a multi-tier, edge-cloud infrastructure where deep learning tasks are extracted from the AR application pipeline and offloaded to nearby GPU-powered edge devices. Our aim is to minimize the latency experienced by end-users and the energy costs on mobile devices. Our multifaceted evaluation, based on benchmarked performance of AR tasks, shows the efficacy of our solution. Overall, Nimbus reduces the task latency by ~4x and the energy consumption by ~77% for real-time object detection in AR applications. We also benchmark three variants of our offloading algorithm, disclosing the trade-off of centralized versus distributed execution.

Context-Aware Offloading for IoT Application using Fog-Cloud Computing

It is difficult to run delay-sensitive applications and the cloud simultaneously due to performance metrics such as latency, energy consumption, bandwidth, and response time exceeding threshold levels. This is the case even though advanced networks and technologies are being used. The middleware layer of the Internet of Things (IoT) architecture appears to be a promising solution that could be used to deal with these issues while still meeting the need for high task offloading criterion. The research that is being proposed recommends implementing Fog Computing (FC) as smart gateway in middleware so that it can provide services the edge of the networks. Applications that are sensitive to delays would then be able to be provided in an efficient manner as a result of this. A smart gateway is proposed as solution for taking the offloading decision based on the context of data, which offers a hybrid approach in order to make decisions regarding offloading that are efficient and effective. A solution that uses machine-learning reasoning techniques to make offloading decisions, in multiple fog-based cloud environments. Feature selection, and classification are used as a learning method and are also ensembled as hybrid logistic regression-based learning to provide the best offloading solution. It works by learning the contextual information of data and identify the cases to make the decision of offloading. The proposed model offers a solution that is both energy and time efficient, with an overall accuracy of approximately 80 percent. With the proposed intelligent offloading approach, it is expected that Internet of Things applications will be able to meet the requirement for low response time and other performance characteristics.

A3C-based load-balancing solution for computation offloading in SDN-enabled vehicular edge computing networks

A3C-based load-balancing solution for computation offloading in SDN-enabled vehicular edge computing networks

Joint Task Offloading, Resource Allocation, and Load-Balancing Optimization in Multi-UAV-Aided MEC Systems

Due to their limited computation capabilities and battery life, Internet of Things (IoT) networks face significant challenges in executing delay-sensitive and computation-intensive mobile applications and services. Therefore, the Unmanned Aerial Vehicle (UAV) mobile edge computing (MEC) paradigm offers low latency communication, computation, and storage capabilities, which makes it an attractive way to mitigate these limitations by offloading them. Nevertheless, the majority of the offloading schemes let IoT devices send their intensive tasks to the connected edge server, which predictably limits the performance gain due to overload. Therefore, in this paper, besides integrating task offloading and load balancing, we study the resource allocation problem for multi-tier UAV-aided MEC systems. First, an efficient load-balancing algorithm is designed for optimizing the load among ground MEC servers through the handover process as well as hovering UAVs over the crowded areas which are still loaded due to the fixed location of the ground base stations server (GBSs). Moreover, we formulate the joint task offloading, load balancing, and resource allocation as an integer problem to minimize the system cost. Furthermore, an efficient task offloading algorithm based on deep reinforcement learning techniques is proposed to derive the offloading solution. Finally, the experimental results show that the proposed approach not only has a fast convergence performance but also has a significantly lower system cost when compared to the benchmark approaches.

A Multihop Task Offloading Decision Model in MEC-Enabled Internet of Vehicles

As a new network technology, mobile-edge computing (MEC) combined with the Internet of Vehicles (IoV) can effectively improve the efficiency of task computing and offloading. However, the power of edge computing will be severely limited to the areas with poor MEC server coverage. Furthermore, there are a number of peripheral vehicles with temporarily idle computing resources on the road, so how to put the resources of these vehicles into use becomes the primary issue to be considered. In this article, a distributed multihop task offloading decision model for task execution efficiency is developed, which mainly consists of two parts: 1) a candidate vehicle selection mechanism for screening the neighboring vehicles that can participate in offloading and 2) a task offloading decision algorithm for obtaining the task offloading solution. Considering the impact of different hop and wireless communication ranges on communication ranges on task completion in a generic scenario, we introduce the hop count <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$k$ </tex-math></inline-formula> and select the neighboring vehicles in the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$k$ </tex-math></inline-formula> -hop wireless communication range as the candidate vehicles. Then, the problem of offloading is modeled as a generalized allocation model with constraints which is solved by the greedy algorithm and discrete bat algorithm, respectively. The results show that compared with the scheme in which the task vehicle randomly selects the neighboring vehicles to offload and the scheme that all tasks are completed locally, the offloading scheme in which all tasks are completed under the greedy algorithm or bat-based algorithm has advantages in time delay performance in terms of different task number, task required computation power, and task size environment. Besides, this article also explores the influence of hop count <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$k$ </tex-math></inline-formula> on the results when selecting candidate vehicles from the neighboring vehicles within the range of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$k$ </tex-math></inline-formula> hop. The results show that the increase of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$k$ </tex-math></inline-formula> will also increase the number of candidate vehicles, which makes the time delay lower. Under the parameters set in this article, the time delay required for the greedy algorithm offloading scheme to complete all tasks is a lower bound on the time delay of the bat algorithm scheme. The greedy algorithm scheme reduces latency by 0.2–2.4 s compared to the scheme where tasks are all completed locally, and it reduces latency by 0.16–2.3 s compared to the random offloading scheme.