Faculty Publications

Now showing 1 - 4 of 4

Containerized deployment of micro-services in fog devices: a reinforcement learning-based approach
(Springer, 2022) Nath, S.B.; Chattopadhyay, S.; Karmakar, R.; Addya, S.K.; Chakraborty, S.; Ghosh, S.K.
The real power of fog computing comes when deployed under a smart environment, where the raw data sensed by the Internet of Things (IoT) devices should not cross the data boundary to preserve the privacy of the environment, yet a fast computation and the processing of the data is required. Devices like home network gateway, WiFi access points or core network switches can work as a fog device in such scenarios as its computing resources can be leveraged by the applications for data processing. However, these devices have their primary workload (like packet forwarding in a router/switch) that is time-varying and often generates spikes in the resource demand when bandwidth-hungry end-user applications, are started. In this paper, we propose pick–test–choose, a dynamic micro-service deployment and execution model that considers such time-varying primary workloads and workload spikes in the fog nodes. The proposed mechanism uses a reinforcement learning mechanism, Bayesian optimization, to decide the target fog node for an application micro-service based on its prior observation of the system’s states. We implement PTC in a testbed setup and evaluate its performance. We observe that PTC performs better than four other baseline models for micro-service offloading in a fog computing framework. In the experiment with an optical character recognition service, the proposed PTC gives average response time in the range of 9.71 sec–50 sec, which is better than Foglets (24.21 sec–80.35 sec), first-fit (16.74 sec–88 sec), best-fit (11.48 sec–57.39 sec) and mobility-based method (12 sec–53 sec). A further scalability study with an emulated setup over Amazon EC2 further confirms the superiority of PTC over other baselines. © 2021, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.
ESMA: Towards elevating system happiness in a decentralized serverless edge computing framework
(Academic Press Inc., 2024) Datta, S.; Addya, S.K.; Ghosh, S.K.
Due to the rapid growth in the adoption of numerous technologies, such as smartphones and the Internet of Things (IoT), edge and serverless computing have started gaining momentum in today's computing infrastructure. It has led to the production of huge amounts of data and has also resulted in increased network traffic, which if not managed well can cause network congestion. To address this and maintain the quality of service (QoS), in this work, a novel dispatch (destination selection) algorithm called Egalitarian Stable Matching Algorithm (ESMA) for faster data processing has been developed while also considering the best use of server resources in a decentralized Serverless-Edge environment. This will allow us to effectively utilize the enormous volumes of data that are generated. The proposed algorithm has been able to achieve lower overall dissatisfaction scores for the entire system. Individually, the client's happiness as well as the server's happiness have improved over the baseline. Moreover, there has been a drop of 25.7% in the total execution time and the total network resources consumed are lower as compared to the baseline algorithm as well as random-allocation algorithm. © 2023 Elsevier Inc.
FASE: fast deployment for dependent applications in serverless environments
(Springer, 2024) Saha, R.; Satpathy, A.; Addya, S.K.
Function-as-a-service has reduced the user burden by allowing cloud service providers to overtake operational activities such as resource allocation, service deployment, auto-scaling, and load-balancing, to name a few. The users are only responsible for developing the business logic through event-triggered functions catering to an application. Although FaaS brings about multiple user benefits, a typical challenge in this context is the time incurred in the environmental setup of the containers on which the functions execute, often referred to as the cold-start time leading to delayed execution and quality-of-service violations. This paper presents an efficient scheduling strategy FASE that uses a finite-sized warm pool to facilitate the instantaneous execution of functions on pre-warmed containers. Test-bed evaluations over AWS Lambda confirm that FASE achieves a 40% reduction in the average cold-start time and 1.29× speedup compared to the baselines. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2023.
Delay-aware partial task offloading using multicriteria decision model in IoT–fog–cloud networks
(Academic Press, 2025) S.a, S.; E, M.; Addya, S.K.; Rahman, S.; Pal, S.; Karmakar, C.
Fog computing plays a prominent role in offloading computational tasks in heterogeneous environments since it provides less service delay than traditional cloud computing. The Internet of Things (IoT) devices cannot handle complex tasks due to less battery power, storage and computational capability. Full offloading has issues in providing efficient computation delay due to more response time and transmission cost. A suitable solution to overcome this problem is to partition the tasks into splittable subtasks. Considering multi-criteria decision parameters like processing efficiency and deadline helps to achieve efficient resource allocation and task assignment. The matching theory is applied to map task nodes to heterogeneous fog nodes and VMs for stability. Compared to baseline algorithms, proposed algorithms like Resource Allocation based on Processing Efficiency (RABP) and Task Assignment Based on Completion Time (TAC) are efficient enough to provide reasonable service delay and discard the non-beneficial tasks, i.e., tasks that do not execute within the deadline. © 2025 The Authors

Faculty Publications

Browse

Filters

Settings

Sort By

Results per page

Search Results