To overcome vendor lock-in and reliability issues in single-cloud deployments, organizations increasingly adopt multi-cloud environments. However, scheduling Spark workflows across heterogeneous clouds under simultaneous deadline and budget constraints remains challenging due to resource diversity, variable pricing, and cross-cloud data transfers. We propose the Deadline Budget Spark Workflow Scheduling to Multi-Cloud (DB-SWSMC) algorithm, a novel scheduling algorithm combining heuristic initialization with simulated annealing optimization to: (1) efficiently allocate resources while balancing cost-time tradeoffs, (2) handle intra/inter-cloud data dependencies, and (3) rigorously enforce constraints. Evaluations across five workflows and compared against existing algorithms (HBDCWS, DBCS, and BDHEFT). Experimental results demonstrate that DB-SWSMC outperforms existing algorithms by 20-40% in cost efficiency and 15-80% in success rates, especially under tight budget and deadline constraints.

The Internet of Things (IoT) is becoming integral to our daily lives, facilitating data collection and analysis for informed decision-making as devices generate vast amounts of data. The transition from Mobile Cloud Computing (MCC) to Mobile Edge Computing (MEC) is increasingly favored for its benefits, including reduced communication latency and efficient bandwidth utilization; however, offloading tasks to the MEC encounters challenges related to data privacy and energy consumption. This study presents an advanced Deep Reinforcement Learning (DRL) based security-aware data offloading and resource allocation model for industrial IoT devices that prioritizes security while effectively managing their computational and radio resources. The model aims to minimize computation latency and energy consumption, which we address using a deep learning optimization strategy. Furthermore, an AES-based security layer is integrated to meet data security needs. Experimental results show that our model significantly reduces offloading overhead compared to both local execution and full offloading, while also demonstrating exceptional scalability for large-scale IoT deployments.

Microservices are being used more and more in the development of cloud-based applications. Using containers, microservice instances can be made to be easier to scale and keep up to date. In order to address the necessity of guaranteeing diverse quality of service (QoS) requirements, the scheduling of microservice workflows in mobile edge computing poses a significant challenge that requires attention and resolution. In this research, a heuristic method called RWSMS is introduced, the system aims to achieve this objective while also ensuring that deadline and reliability criteria are met. This study presents a proposed scheduling strategy that incorporates RWSMS algorithms to minimize the cost of workflow execution, while simultaneously ensuring that the user-defined deadline is met and reliability is assured. Additionally, the RWSMS system incorporates a resource adjustment approach in order to optimize resource consumption. By conducting a series of comprehensive experiments using different real-world workflow applications, the effectiveness and efficiency of RWSMS are evaluated and compared with existing algorithms. The results confirm that RWSMS successfully achieves lower execution costs while meeting the required deadlines and ensuring reliability.

To fulfill their expanding computational demands, businesses are using cloud computing more and more these days. However, cloud systems alone may not always suffice. Consequently, multi-cloud systems, which provide more scalable storage and computing resources, are becoming more popular. This paper focuses on scheduling Spark workflow tasks in a multi-cloud environment. It addresses the challenges posed by different pricing models, dynamic resource provisioning, inter and intra transmission time, and the instability of resource performance. To tackle these issues, in this work, we propose a heuristic-based solution that considers factors such as VM instances, precedence constraints, transmission times, and the impact of performance uncertainty aiming to minimize rental costs while ensuring that workflow deadlines are met. The results show that the proposed method is effective in scheduling Spark workflow tasks in a multi-cloud environment while considering performance uncertainty.

Serverless applications based on microservices typically comprise dozens or hundreds of loosely coupled functions. Each request in these applications triggers a function chain, which invokes a subset of functions. However, the invocation paths may not be predetermined in dynamic function chains. In addition, there are multiple memory configurations for each function in a chain. Different memory configuration combinations bring different execution times and costs. The configuration space grows exponentially as the length of the function chain increases. Therefore, it is challenging to determine the memory configurations for functions in a dynamic function chain to minimize the execution time and cost with an uncertain invocation path and a large configuration space. In this paper, a memory configuration problem of linear dynamic function chains is investigated to minimize the cost while meeting a specified SLO (service level objective). RLMC (Reinforcement Learning-based Memory Configuration Algorithm) is adopted to make memory configuration decisions dynamically. States, actions, and rewards are specially designed for the problem under study. In addition, a punishment factor adjustment strategy is developed to accommodate different SLOs. The proposed algorithm is evaluated and compared to existing algorithms over a comprehensive set of randomly generated serverless workflow applications. Experimental results demonstrate that RLMC significantly reduces the cost of dynamic function chains while meeting SLOs and outperforms other algorithms.

These days, the usage of cloud computing services for different applications has been growing progressively. The applications, including business, commerce, healthcare, and others, require additional computation capabilities for their executions. To fulfil their expanding computational demands, cloud computing offers a pay-as-you-go billing model to run these applications cost-effectively. However, due to the complex requirements of these applications, more than one cloud system is required because single-cloud solutions are often limited by resource constraints, such as inadequate storage and computing power, as well as single-point failures that can compromise the integrity of the entire application. Consequently, multi-cloud strategies, which provide more scalable storage and computing resources, are becoming increasingly popular. However, the multi-cloud landscape consists of many cloud providers, and effectively managing workflow scheduling presents a significant hurdle in this dynamic environment. This paper focuses on scheduling Spark workflow tasks in multi-cloud networks. It addresses the challenges posed by different pricing models, dynamic resource provisioning, inter- and intra-transmission time, and the instability of resource performance. To solve these challenges, we propose a novel heuristic-based approach that considers different constraints such as VM instances heterogeneity, priority constraints, transmission times, and the impact of performance uncertainty. The goal is to schedule all tasks on virtual machines (VMs) with rental costs as low as possible while meeting workflow deadlines. The simulation results show that the proposed method effectively schedules Spark workflow tasks in multi-cloud networks, improving the scheduling performance by 50% compared to existing approaches.