• 한국정보통신설비학회:학술대회논문집
• /
• 2006.08a
• /
• pp.77-81
• /
• 2006

Because of the rapid increasing of network user, there are some problems to tolerate the network overhead. Recently, the research and technology of the user-level for high performance and low latency than TCP/IP which relied upon the kernel for processing the messages. For example, there is an Infiniband technology. The Infiniband Trade Association (IBTA) has been proposed as an industry standard for both communication between processing node and I/O devices and for inter-processor communication. It replaces the traditional bus-based interconnect with a switch-based network for connecting processing node and I/O devices. Also Infiniband uses RDMA (Remote DMA) for low latency of CPU and OS to communicate between Remote nodes. In this paper, we develop the SRP (SCSI RDMA Protocol) which is Storage Access Protocol on Infiniband network. And will compare to FC (Fibre Channle) based I-SCSI (Internet SCSI) that it is used to access storage on Etherent Fabric.

• International journal of advanced smart convergence
• /
• v.13 no.3
• /
• pp.143-149
• /
• 2024

As the size of data and models in machine learning training continues to grow, training on a single server is becoming increasingly challenging. Consequently, the importance of distributed machine learning, which distributes computational loads across multiple machines, is becoming more prominent. However, several unresolved issues remain regarding the performance enhancement of distributed machine learning, including communication overhead, inter-node synchronization challenges, data imbalance and bias, as well as resource management and scheduling. In this paper, we propose ParamHub, which utilizes orchestration to accelerate training speed. This system monitors the performance of each node after the first iteration and reallocates resources to slow nodes, thereby speeding up the training process. This approach ensures that resources are appropriately allocated to nodes in need, maximizing the overall efficiency of resource utilization and enabling all nodes to perform tasks uniformly, resulting in a faster training speed overall. Furthermore, this method enhances the system's scalability and flexibility, allowing for effective application in clusters of various sizes.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.2
• /
• pp.378-384
• /
• 2008

The sensor network for the health care application service has the man or movable object as the main sensing object. In order to support inter-node interaction by the movement of such sensing objects, the node's dynamic function modification, dynamic self-configuration and energy efficiency must be considered. In this paper, the Agilla model which supports the dynamic function modification through the agent migration between nodes and LEACH protocol which guarantees the dynamic self-configuration and energy efficiency through the configuration of inter-node hierarchical cluster configuration are analyzed. Based on the results of the analysis, the Mobility Agent Middleware which supports the dynamic function modification between nodes is designed, and LEACH_Mobile protocol which guarantees the node nobility as the weakness of the existing LEACH protocol is suggested. Also, the routing module which supports the LEACH_Mobile protocol is designed and the interface for conjunction with Mobility Agent Middleware is designed. Then, it is definitely increase performance which un mobility node of transfer data rate through LEACH_Mobile protocol of simulation result.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.5 no.8
• /
• pp.1388-1403
• /
• 2011

A vehicular ad-hoc network (VANET) consists of vehicles that form a network without any additional infrastructure, thus allowing the vehicles to communicate with each other. VANETs have unique characteristics, including high node mobility and rapidly changing network topology. Because of these characteristics, routing algorithms based on greedy forwarding such as greedy perimeter stateless routing (GPSR) are known to be very suitable for a VANET. However, greedy forwarding just selects the node nearest to the destination node as a relay node within its transmission range. This increases the possibility of a local maximum and link loss because of the high mobility of vehicles and the road characteristics in urban areas. Therefore, this paper proposes a reliability-improving position-based routing (RIPR) algorithm to solve those problems. The RIPR algorithm predicts the positions, velocities, and moving directions of vehicles after receiving beacon messages, and estimates information about road characteristics to select the relay node. Thus, it can reduce the possibility of getting a local maximum and link breakage. Simulation results using ns-2 revealed that the proposed routing protocol performs much better than the existing routing protocols based on greedy forwarding.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.48 no.2
• /
• pp.14-20
• /
• 2011

In mobile ad hoc networks where the communication nodes are moving around, one may perform dynamic routing that can increase the total communication throughput of the network, by determining the ranks of pairs of nodes according to their communication throughput. The balanced Howell rotation is a tournament design scheme for bridge games. This paper explains that the balanced Howell rotation can be applied to enhance the overall communication throughput of mobile ad hoc networks, and proposes and proves the condition under which the balanced Howell rotations may exist.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.11
• /
• pp.5338-5359
• /
• 2017

In-band full-duplex (IBFD) wireless communication supports symmetric dual transmission between two nodes and asymmetric dual transmission among three nodes, which allows improved throughput for distributed IBFD wireless networks. However, inter-node interference (INI) can affect desired packet reception in the downlink of three-node topology. The current Half-duplex (HD) medium access control (MAC) mechanism RTS/CTS is unable to establish an asymmetric dual link and consequently to suppress INI. In this paper, we propose a medium access control mechanism for use in distributed IBFD wireless networks, FD-DMAC (Full-Duplex Distributed MAC). In this approach, communication nodes only require single channel access to establish symmetric or asymmetric dual link, and we fully consider the two transmission modes of asymmetric dual link. Through FD-DMAC medium access, the neighbors of communication nodes can clearly know network transmission status, which will provide other opportunities of asymmetric IBFD dual communication and solve hidden node problem. Additionally, we leverage FD-DMAC to transmit received power information. This approach can assist communication nodes to adjust transmit powers and suppress INI. Finally, we give a theoretical analysis of network performance using a discrete-time Markov model. The numerical results show that FD-DMAC achieves a significant improvement over RTS/CTS in terms of throughput and delay.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.12
• /
• pp.1639-1645
• /
• 2020

In this paper, we propose a virtual full-duplex relaying scheme based on data ferrying using two unmanned aerial vehicles (UAVs). By utilizing high mobility of two UAVs, the proposed relaying scheme can make each UAV communicate with a source node or a destination node when the UAV is near the source node or the destination node. The proposed relaying scheme can overcome the performance limitations of the half-duplex relaying and the implementation constraints of the full-duplex relaying. In addition, we propose an algorithm to shift the center point of two UAVs' trajectory with consideration of inter-relay interference (IRI). We show the simulation results of our proposed trajectory's center point shift algorithm. From the simulation results, it is shown that our proposed relaying scheme can achieve higher end-to-end spectral efficiency (SE) than the conventional static relaying scheme.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.1
• /
• pp.107-115
• /
• 2009

The cryptological key which is used in WiMAX environment is used at regular intervals by mobile nodes (laptop computer, PDA, cell-phone) which is in the range of base station coverage. But it is very weak at local attack like man-in-the-middle when the mobile node is off the range of base station or enters into the range to communicate with base station because the communication section is activated wirelessly. This paper proposes a distribution key building protocol which can reuse security key used by nodes to reduce cryptological security attack danger and communication overhead which occurs when mobile node tries to communicate with base station. The proposed distribution key establishing protocol can reduce overhead which occurs between base station and mobile node through key reusing which occurs during the communication process and also, makes security better than IEEE 802.16 standard by creating shared key which is required for inter-certification through the random number which node itself creates.

• International Journal of Computer Science & Network Security
• /
• v.23 no.1
• /
• pp.153-163
• /
• 2023

This study is situated in the context of intelligent transport systems, where in-vehicle devices assist drivers to avoid accidents and therefore improve road safety. The vehicles present in a given area form an ad' hoc network of vehicles called vehicular ad' hoc network. In this type of network, the nodes are mobile vehicles and the messages exchanged are messages to warn about obstacles that may hinder the correct driving. Node mobilities make it impossible for inter-node communication to be end-to-end. Recognizing this characteristic has led to delay-tolerant vehicular networks. Embedded devices have small buffers (memory) to hold messages that a node needs to transmit when no other node is within its visibility range for transmission. The performance of a vehicular delay-tolerant network is closely tied to the successful management of the nodes' transit buffer. In this paper, we propose a message transit buffer management model for nodes in vehicular delay tolerant networks. This model consists in setting up, on the one hand, a policy of dropping messages from the buffer when the buffer is full and must receive a new message. This drop policy is based on the concept of intermediate node to destination, queues and priority class of service. It is also based on the properties of the message (size, weight, number of hops, number of replications, remaining time-to-live, etc.). On the other hand, the model defines the policy for selecting the message to be transmitted. The proposed model was evaluated with the ONE opportunistic network simulator based on a 4000m x 4000m area of downtown Bouaké in Côte d'Ivoire. The map data were imported using the Open Street Map tool. The results obtained show that our model improves the delivery ratio of security alert messages, reduces their delivery delay and network overload compared to the existing model. This improvement in communication within a network of vehicles can contribute to the improvement of road safety.

• Journal of KIISE:Information Networking
• /
• v.31 no.4
• /
• pp.353-362
• /
• 2004

Mobile Ad-Hoc network tends to expose scarce computing resources and various security threats because all traffics are carried in air along with no central management authority. To provide secure communication and save communication overhead, a scheme is inevitable to serurely establish session keys. However, most of key establishment methods for Ad-Hoc network focus on the distribution of a group key to all hosts and/or the efficient public key management. In this paper, a secure and efficient scheme is proposed to establish a session key between two Ad-Hoc nodes. The proposed scheme makes use of the secret sharing mechanism and the Diffie-Hellman key exchange method. For secure intra-cluster communication, each member node establishes session keys with its clusterhead, after mutual authentication using the secret shares. For inter-cluster communication, each node establishes session keys with its correspondent node using the public key and Diffie-Hellman key exchange method. The simulation results prove that the proposed scheme is more secure and efficient than that of the Clusterhead Authentication Based Method(1).