The System Architecture and Methods for Efficient Resource-Saving Scheduling in the Fog †
Abstract
:1. Introduction
- The distances between fog nodes affect the time of data exchange between the fog nodes. Time consumption leads to the impossibility of meeting the QoS requirements and also worsens the reliability functions of the fog nodes because of possible overloading;
- The scheduling problem itself must be solved as soon as possible because of the need to maintain the QoE and QoS requirements;
- Finally, the forming of a resource-saving schedule requires the nodes’ workload history, at least when the log of the device resource states changes.
- The search space for the optimization problem solving must be formed on the basis of available fog nodes;
- The workload history must be used for the schedule formation.
- Resource-saving scheduling problem formalization and analysis;
- A state-of-the-art analysis in the field of scheduling in the fog, resource saving, and distributed ledger;
- A comparison of the approaches to providing CFB with the appropriate data;
- System architecture and basic functional method development.
2. The Resource-Saving Scheduling Problem and Its Analysis
- D is the node workload;
- k is the coefficient of node temperature increase depending on the current workload, and
- is the moment of assignment of task j to the node i.
- The search space, which is the set of nodes that are available for the task assignment;
- The workload story for the fog nodes, which is needed to estimate the reliability function values, as it is the integral part of the overall objective function.
3. State-of-the-Art Analysis
- The computational resources, node failure rate and reliability function connected to the nodes workload;
- The particular scheduling problem-solving under the time constraints and the uncertainty of the resources;
- The distributed ledger technologies and their application to the node information provisioning.
4. A Comparison of the Approaches to CFB Data Provisioning and the Approach Choice for System Design
- A search space for the optimization of problem-solving;
- Information about the workload history of the fog nodes.
- There are three node types in the system: user (edge) nodes, fog nodes, and CFB nodes. User nodes are the sources of the tasks and data, fog nodes are supposed to participate in the tasks and data processing, and CFB nodes produce schedules and distribute computational workload through the fog nodes and cloud.
- The need to deal with the local copies of the workload history and resource states forms the main requirement to the distributed ledger functioning: the data on the resource state change of the fog node must be placed into the ledger in an order of events.
- The additional requirement for the fog node information collecting and disseminating is that information about the node resource state change must be disseminated through the network as soon as possible.
- In the case of assigning the fog nodes to the CFB, there is a need to detect CFB failure and to restore the information provisioning as soon as possible.
5. Development of the System Architecture and Basic Functional Methods
- Provides the node state registration;
- Provides the storage of the device-broker assignment information;
- Provides the fog node–broker interaction;
- Provides the procedures of receiving and processing the computational tasks as planned by a fog broker.
- Interaction with other brokers;
- Interaction with the fog nodes;
- Interaction with the ledger replicas and the providing of its functioning;
- Scheduling problem-solving.
5.1. Adding a New Broker Node to the Broker’s Network
- 1.
- The broker agent initializes.
- 2.
- Broker agent sends a request to the neighbour nodes about their participation in the process of information delivery for the resource-saving scheduling problem-solving. If some nodes are found, then the following occurs:
- Request of the list of fog brokers from the neighbour fog node.
- Sending of a request of the ledger copy to the nearest fog broker node.
- Addition of the new own blockchain to the block lattice, which is assigned to the new device in the fog layer.
- 3.
- If there is no devices with ledger copies, then, the current fog broker is the first ledger copy.
- 4.
- Creation the first ledger copy with the blockchain assigned to the broker node.
- 5.
- Creation of the new list of brokers, each with its identifier being added to the list.
- 6.
- Implementation of the search through the network cluster to gather the information about the fog nodes’ current states.
- 7.
- Addition of new blockchains to the ledger and their being assigned with the new-found nodes.
- 8.
- Sending of the broker ID to the active fog nodes.
5.2. Broker Failure
- The fog node sends the state data to some of the nearest fog brokers. In the case of broker failure, no data are lost, yet the question of data duplication emerges.
- The fog node sends state data to the nearest fog broker.
5.3. Functioning Stage
- 1.
- The leader sends “heartbeat” messages to its followers.
- 2.
- The follower sends the proof of the message’s receipt.
5.4. Follower Failure
- 1.
- If the current leader has not received the heartbeat message, then;
- 2.
- The leader searches its ledger copy for the fog nodes assigned to the failed broker;
- 3.
- The leader searches the fog broker, which is nearest to the fog nodes assigned to the failed broker;
- 4.
- The request of the new fog broker setup is sent to the fog nodes;
- 5.
- The new assigned broker sends the request for fog node states to the fog nodes;
- 6.
- States are put into the ledger copy;
- 7.
- The state of the failed fog broker is put into the ledger as a state with the full utilization (no resources available);
- 8.
- New data are replicated through all the brokers.
5.5. Leader Failure
- 1.
- Followers wait for the “heartbeat” from the leader node;
- 2.
- If there is no “heartbeat”, the new leader is elected, for instance, by means of some simple procedure (round robin);
- 3.
- The search of the new broker, which is near the nodes of the failed broker, is conducted;
- 4.
- The request of the new fog broker setup is sent to the fog nodes;
- 5.
- The new fog broker sends the state request to the new fog nodes;
- 6.
- The received states are put into the ledger copy;
- 7.
- The state of the failed broker is put to the ledger as the state of full utilization (the lack of available resources);
- 8.
- New data are replicated to the ledger copies.
5.6. New Fog Device Addition
- 1.
- Within the new fog node emergence, the fog node agent sends its identifier into the network.
- 2.
- The first broker receives this message, sends the confirmation to the new node, and blocks it for the other brokers. From this moment the fog node is presupposed to be assigned to the current broker.
- 3.
- The assigned broker requests the node state information and puts it into the ledger.
5.7. Fog Node Failure
6. Conclusions
- Search space formation;
- Resource-saving scheduling problem-solving.
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Xu, J.; Hao, X.; Zhang, R.; Sun, X. A Method Based on the Combination of Laxity and Ant Colony System for Cloud-Fog Task Scheduling. IEEE Access 2019, 7, 116218–116226. [Google Scholar] [CrossRef]
- Alenizi, F.; Rana, O. Fog Computing: Towards Dynamically Controlling the Offloading Threshold and Managing Fog Resources in Online Dynamic Systems. Preprints.org 2021. [Google Scholar] [CrossRef]
- Giang, N.; Lea, R. Developing applications in large scale, dynamic fog computing: A case study. Softw. Pract. Exp. 2020, 50, 519–532. [Google Scholar] [CrossRef]
- Zhiglyavsky, A.; Zhilinskas, A. A Global Extremum Search Methods; Nauka: Moscow, Russia, 1991; p. 247. [Google Scholar]
- Klimenko, A.; Melnik, E. Information and Control Systems with Distributed Ledger Usage: A Reliability Issue. In Artificial Intelligence in Intelligent Systems: Proceedings of 10th Computer Science On-Line Conference 2021; Springer International Publishing: New York, NY, USA, 2021. [Google Scholar] [CrossRef]
- Klimenko, A.; Kalyaev, I. A Technique to Provide an Efficient System Recovery in the Fog- and Edge-Environments of Robotic Systems. In Proceedings of the Interactive Collaborative Robotics: 6th International Conference, St. Petersburg, Russia, 27–30 September 2021. [Google Scholar] [CrossRef]
- Korovin, I.; Melnik, E.; Klimenko, A. The Fog-Computing Based Reliability Enhancement in the Robot Swarm. In Proceedings of the Interactive Collaborative Robotics: 4th International Conference, ICR 2019, Istanbul, Turkey, 20–25 August 2019. [Google Scholar] [CrossRef]
- Blazewicz, J.; Moseley, B.; Pesch, E.; Trystram, D.; Zhang, G. New Perspectives in Scheduling Theory. J. Sched. 2021, 24, 161–169. [Google Scholar] [CrossRef]
- Klimenko, A.; Safronenkova, I. An Ontology-Based Approach to the Workload Distribution Problem Solving in Fog-Computing Environment. In Artificial Intelligence Methods in Intelligent Algorithms: Proceedings of 8th Computer Science On-line Conference 2019; Springer International Publishing: New York, NY, USA, 2019. [Google Scholar] [CrossRef]
- Wu, D.; Ansari, N. A Cooperative Computing Strategy for Blockchain-Secured Fog Computing. IEEE Internet Things J. 2020, 7, 6603–6609. [Google Scholar] [CrossRef]
- Alam, T. IoT-Fog-Blockchain Framework: Opportunities and Challenges. Int. Fog Comput. 2020, 3, 1–20. [Google Scholar] [CrossRef]
- Son, B.; Lee, J.; Jang, H. A Scalable IoT Protocol via an Efficient DAG-Based Distributed Ledger Consensus. Sustainability 2020, 12, 1529. [Google Scholar] [CrossRef]
- Kalyaev, I.; Melnik, E.; Klimenko, A. Distributed Ledger Based Workload Logging in the Robot Swarm. In Proceedings of the Interactive Collaborative Robotics: 4th International Conference, ICR 2019, Istanbul, Turkey, 20–25 August 2019. [Google Scholar] [CrossRef]
- Masood, F.; Faridi, A. An Overview of Distributed Ledger Technology and its Applications. Int. J. Comput. Sci. Eng. 2018, 6, 422–427. [Google Scholar] [CrossRef]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Klimenko, A. The System Architecture and Methods for Efficient Resource-Saving Scheduling in the Fog. Eng. Proc. 2023, 33, 9. https://doi.org/10.3390/engproc2023033009
Klimenko A. The System Architecture and Methods for Efficient Resource-Saving Scheduling in the Fog. Engineering Proceedings. 2023; 33(1):9. https://doi.org/10.3390/engproc2023033009
Chicago/Turabian StyleKlimenko, Anna. 2023. "The System Architecture and Methods for Efficient Resource-Saving Scheduling in the Fog" Engineering Proceedings 33, no. 1: 9. https://doi.org/10.3390/engproc2023033009