Decentralized intelligent routing algorithm in distributed database cluster

In the current communication technology, optical technology has been applied to the network to obtain optical network technology. Among them, optical network technology is optical wavelength division multiplexing (WDM), which can play a larger transmission capacity under lower energy consumption. Further breakthroughs in intelligent optical networks require improvements in routing issues. In this study, firstly, the optical network architecture is analyzed, including wavelength division multiplexing optical network and elastic optical network. Then, the routing problem in optical networks is analyzed, and the main factors affecting the routing problem are extracted. On the basis of studying the energy consumption characteristics of data centers and WDM optical networks, and considering the characteristics of cloud service configuration, evolutionary game theory and optical bypass theory are introduced to obtain an intelligent routing algorithm for cloud computing based on optical networks, and energy consumption tests are carried out on data transmission and processing. In order to reduce the overall energy consumption, the use of IP routers is reduced, and the idle data servers are shut down. Then, it is found that the total energy consumption increases slowly at different times. The energy consumption of evolutionary game theory is compared. Compared with non-evolutionary game theory, the optimized intelligent routing algorithm makes the energy consumption more stable, while reducing the use of servers can further reduce the good expenditure. The proposed algorithm is oriented to optical network, which solves the problem of low overall utilization of network resources and improves the service quality of cloud services.

Download Full-text

An Intelligent Routing Algorithm in Wireless Sensor Networks Based on Reinforcement Learning

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.678.487 ◽

2014 ◽

Vol 678 ◽

pp. 487-493 ◽

Cited By ~ 2

Author(s):

Wen Jing Guo ◽

Cai Rong Yan ◽

Yang Lan Gan ◽

Ting Lu

Keyword(s):

Wireless Sensor Networks ◽

Reinforcement Learning ◽

Sensor Networks ◽

Network Lifetime ◽

Routing Algorithm ◽

Wireless Sensor ◽

Energy Aware ◽

Hop Count ◽

Packet Delivery ◽

Intelligent Routing

Lifetime enhancement has been a hot issue in Wireless Sensor Networks (WSNs). To prolong the network lifetime of WSNs, this paper proposes an intelligent routing algorithm named RLLO. RLLO makes uses of the superiority of reinforcement learning (RL) and considers residual energy and hop count to define the reward function. It is to uniformly distribute the energy consumption and improve the packet delivery without additional cost. This proposed algorithm has been compared with Energy Aware Routing (EAR) and improved EAR (I-EAR). Simulation results show that RLLO gains a significant improvement in terms of network lifetime and packet delivery over these two algorithms.

Download Full-text

DRSIR: A Deep Reinforcement Learning Approach for Routing in Software-Defined Networking

10.36227/techrxiv.14501604 ◽

2021 ◽

Author(s):

Daniela Casas Velasco ◽

Oscar Mauricio Caicedo Rendon ◽

Nelson Luis Saldanha da Fonseca

Keyword(s):

Reinforcement Learning ◽

Routing Protocols ◽

Routing Algorithm ◽

Routing Algorithms ◽

Limited Information ◽

State Spaces ◽

Viable Solution ◽

Slow Adaptation ◽

Intelligent Routing

Traditional routing protocols employ limited information to make routing decisions which leads to slow adaptation to trafﬁc variability and restricted support to the quality of service requirements of the applications. To address these shortcomings, in previous work, we proposed RSIR, a routing solution based on Reinforcement Learning (RL) in SoftwareDeﬁned Networking (SDN). However, RL-based solutions usually suffer an increase in the learning process when dealing with large action and state spaces. This paper introduces a different routing approach called Deep Reinforcement Learning and SoftwareDeﬁned Networking Intelligent Routing (DRSIR). DRSIR deﬁnes a routing algorithm based on Deep RL (DRL) in SDN that overcomes the limitations of RL-based solutions. DRSIR considers path-state metrics to produce proactive, efﬁcient, and intelligent routing that adapts to dynamic trafﬁc changes. DRSIR was evaluated by emulation using real and synthetic trafﬁc matrices. The results show that this solution outperforms the routing algorithms based on the Dijkstra’s algorithm and RSIR, in relation to stretching (stretch), packet loss, and delay. Moreover, the results obtained demonstrate that DRSIR provides a practical and viable solution for routing in SDN.

Download Full-text

DRSIR: A Deep Reinforcement Learning Approach for Routing in Software-Defined Networking

10.36227/techrxiv.14501604.v1 ◽

2021 ◽

Author(s):

Daniela Casas Velasco ◽

Oscar Mauricio Caicedo Rendon ◽

Nelson Luis Saldanha da Fonseca

Keyword(s):

Reinforcement Learning ◽

Routing Protocols ◽

Routing Algorithm ◽

Routing Algorithms ◽

Limited Information ◽

State Spaces ◽

Viable Solution ◽

Slow Adaptation ◽

Intelligent Routing

Traditional routing protocols employ limited information to make routing decisions which leads to slow adaptation to trafﬁc variability and restricted support to the quality of service requirements of the applications. To address these shortcomings, in previous work, we proposed RSIR, a routing solution based on Reinforcement Learning (RL) in SoftwareDeﬁned Networking (SDN). However, RL-based solutions usually suffer an increase in the learning process when dealing with large action and state spaces. This paper introduces a different routing approach called Deep Reinforcement Learning and SoftwareDeﬁned Networking Intelligent Routing (DRSIR). DRSIR deﬁnes a routing algorithm based on Deep RL (DRL) in SDN that overcomes the limitations of RL-based solutions. DRSIR considers path-state metrics to produce proactive, efﬁcient, and intelligent routing that adapts to dynamic trafﬁc changes. DRSIR was evaluated by emulation using real and synthetic trafﬁc matrices. The results show that this solution outperforms the routing algorithms based on the Dijkstra’s algorithm and RSIR, in relation to stretching (stretch), packet loss, and delay. Moreover, the results obtained demonstrate that DRSIR provides a practical and viable solution for routing in SDN.

Download Full-text

Intelligent Routing Algorithm Using Antnet

International Journal of Trend in Scientific Research and Development ◽

10.31142/ijtsrd18990 ◽

2018 ◽

Vol Volume-3 (Issue-1) ◽

pp. 306-314

Author(s):

Aneke Israel Chinagolum ◽

Chineke Amaechi Hyacenth ◽

Udeh Chukwuma Callistus. W ◽

Keyword(s):

Routing Algorithm ◽

Intelligent Routing

Download Full-text

CLUSTER BASED ROUTING PROTOCOLS IN WSN SENSOR

INFORMATION TECHNOLOGY IN INDUSTRY ◽

10.17762/itii.v9i1.119 ◽

2021 ◽

Vol 9 (1) ◽

pp. 198-206

Author(s):

Kanak Prabha Lila Ramani, Dr. Abhishek Badholia

Keyword(s):

Sensor Network ◽

Network Lifetime ◽

Routing Protocol ◽

Selection Process ◽

Routing Algorithm ◽

Cluster Head ◽

Wireless Sensor ◽

Sleep Mode ◽

Local Monitoring ◽

Intelligent Routing

Over the past few years we can observe the WSNs or Wireless Sensor Network applications in various fields increasing immensely. The energy efficiency, network lifetime and clustering process prime goal is the working network's optimization is the focus of many of the routing algorithm. Keeping in mind the network homogeneity, for network performance reinforcement we suggest instead of single path to use multiple paths. For WSNs, Reinforcement Intelligence Routing Protocol (RIRP)[1]. In the multihop wireless sensor networks an efficient and effective method for security improvements local monitoring has worked well [2]. But taking in consideration the power consumption in the current practice of local monitoring is costly. For ensuring long-lived operations in the sensor network reinforcement intelligent routing protocol is critical [3]. For ensuring both the aspects improvement in security and long-lived operations, the development of mechanism that is effective and incorporated with the Reinforcement protocol is an open problem. With the help of local monitoring to solve this issue, section of the traffic going in and out of its neighbors is supervised by each node to keep a check on any suspicious behavior like unlikely long delays in packet forwarding [4]. To integrate the existing reinforcement protocol of the network and without any niggardly in the consumption of energy in the local monitoring with the help of a protocol [5]. In comparison to other protocols in this protocol the region of instability starts later. At a constant rate the nodes of the RIRP or Reinforcement Intelligent Routing Protocol dies. Few problems such as cluster head selection process, network lifetime and network stability are evaluated and worked in the technique proposed here [6]. To reduce the overload consumption as much as possible the nodes switches in between the active and sleep mode.

Download Full-text