• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.9
• /
• pp.54-62
• /
• 2015

Although NAND flash-based SSD (Solid-State Drive) provides superior performance in comparison to HDD (Hard Disk Drive), it has a major drawback in write endurance. As a result, the lifetime of SSD is determined by the workload and thus it becomes a big challenge in current technology trend of such as the shifting from SLC (Single Level Cell) to MLC (Multi Level cell) and even TLC (Triple Level Cell). Most previous studies have dealt with wear-leveling or improving SSD lifetime regarding hardware architecture. In this paper, we propose the optimal configuration of host I/O stack focusing on file system, I/O scheduler, and link power management using JEDEC enterprise workloads in terms of WAF (Write Amplification Factor) which represents the efficiency perspective of SSD life time especially for host write processing into flash memory. Experimental analysis shows that the optimum configuration of I/O stack for the perspective of SSD lifetime is MinPower-Dead-XFS which prolongs the lifetime of SSD approximately 2.6 times in comparison with MaxPower-Cfq-Ext4, the best performance combination. Though the performance was reduced by 13%, this contributions demonstrates a considerable aspect of SSD lifetime in relation to I/O stack optimization.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.5 no.3
• /
• pp.27-45
• /
• 2010

It is strongly believed that multicast will become one of the most promising services on internet for the next generation. Multicast service can be deployed either on network-layer or application-layer. IP multicast (network-layer multicast) is implemented by network nodes (i.e., routers) and avoids multiple copies of the same datagram on the same link. Despite the conceptual simplicity of IP multicast and its obvious benefits, it has not been widely deployed since there remain many unresolved issues. As an alternative to IP multicast, overlay multicast (application-layer multicast) implements the multicast functionality at end hosts rather than routers. This may require more overall bandwidth than IP multicast because duplicate packets travel the same physical links multiple times, but it provides an inexpensive, deployable method of providing point-to-multipoint group communication. In this paper we develop an efficient method applied greedy algorithm for solving two models of overlay multicast routing protocol that is aimed to construct MDST (Minimum Diameter Spanning Tree : minimum cost path from a source node to all its receivers) and MST (Minimum Spanning Tree : minimum total cost spanning all the members). We also simulate and analyze MDST and MST.

• The KIPS Transactions:PartC
• /
• v.12C no.6 s.102
• /
• pp.933-940
• /
• 2005

As the appearance of very high speed telecommunication network, volume of network, is enlarged and complicated, management of the various network equipments and hosting systems become more complicated and significant. Recently, there have been various researches on network management system that is capable of managing and operating the network environment based on SNMP (Simple Network Management Protocol). SNMP has many advantages, which is easy to implement and has a simple structure. However, as the network structure has become more complicated, it has caused a number of problems like the increase of network load and limit of the network management scope in terms of the network expansion and efficiency. Especially, it needs expensive cost and time for developing a network because many network developers are almost depended manually for developing it till now. In this paper, we propose a framework for network management program that automates the generation of information for network management. The Proposed framework is able to automatically generate a network management program by using information related with equipments which were provided along with the network equipments and SNMP library Thus, we ill make not only the SNMP network structure expansion become easier but also errors maintaining and development time of the network management program were dramatically reduced by using generated network program through our proposed framework.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.11C
• /
• pp.1064-1073
• /
• 2002

In this paper, we design and implement the Wireless Transaction Protocol (WTP) proposed by the Wireless Application Protocol (WAP) forum using a protocol development tool, SDL Development Tool (SDT). And we conduct a comparative performance evaluation of the WTP implementation with other three implementations that are based on different implementation models respectively: the server model, the coroutine model, and the activity-thread model. To implement WTP, we first use Unified Modeling Language (UML) for analyzing the protocol requirement and defining the protocol engine architecture. Next, we use Software Development Language (SDL) to design the protocol engine in details and then generate the WTP implementation automatically with the aid of SDT The code size of the WTP implementation generated by SDT is 62% larger than the other three implementations. However, its throughput and system response time for transaction processing is almost equal to the other three implementations when the number of concurrent clients is less than 3,000. If more than 5,000 concurrent clients tries, the transaction success rate abruptly decreases to 10% and system response time increases to 1,500㎳, due to the increased protocol processing time. But, it comes from the fact that the load overwhelms the capacity of the PC resource used in this experimentation.

• The KIPS Transactions:PartC
• /
• v.14C no.1 s.111
• /
• pp.27-38
• /
• 2007

IPv6 has appeared for solving the address exhaustion of IPv4 and for guaranteeing the problems of security and QoS. It occurs the unexpected new attacks of IPv6 as well as the existing attacks of IPv4 because of the increasing address space to 128bits and the address hierarchies for efficient network management and additions of the new messages between nodes and routers like neighbor discovery and auto address configuration for the various comfortable services. For the successful transition from IPv4 to IPv6, we should get the secure compatibility between IPv4 hosts or routers working based on secure and systematic policy and IPv6. Network manager should design security technologies for efficient management in IPv4/IPv6 co-existence network and IPv6 network and security management framework designation. In this paper, we inspected the characteristics of IPv4 and IPv6, study on security requirement for efficient security management of various attacks, protocol, service in IPv4/IPv6 co-existence and IPv6 network, and finally suggest integrated solution about security vulnerability of IPv6 network in considering of analysis of IPv6 system, host and application, IPv6 characteristics, modified CGA(MCGA).

• Journal of KIISE:Databases
• /
• v.31 no.2
• /
• pp.108-121
• /
• 2004

In this paper, we propose an enhanced snapshot technique that solves performance degradation when snapshot is initiated for the storage cluster system. However, traditional snapshot technique has some limits adapted to large amount storage shared by multi-hosts in the following aspects. As volume size grows, (1) it deteriorates crucially the performance of write operations due to additional disk access to verify COW is performed. (2) Also it increases excessively the blocking time of write operation performed during the snapshot creation time. (3)Finally, it deteriorates the performance of write operations due to additional disk I/O for mapping block caused by the verification of COW. In this paper, we propose an efficient snapshot technique for large amount storage shared by multi-hosts in SAN Environments. We eliminate the blocking time of write operation caused by freezing while a snapshot creation is performing. Also to improve the performance of write operation when snapshot is taken, we introduce First Allocation Bit(FAB) and Snapshot Status Bit(SSB). It improves performance of write operation by reducing an additional disk access to volume disk for getting snapshot mapping block. We design and implement an efficient snapshot technique, while the snapshot deletion time, improve performance by deallocation of COW data block using SSB of original mapping entry without snapshot mapping entry obtained mapping block read from the shared disk.

• 한국벤처창업학회:학술대회논문집
• /
• 2010.08a
• /
• pp.71-89
• /
• 2010

It is strongly believed that multicast will become one of the most promising services on internet for the next generation. Multicast service can be deployed either on network-layer or application-layer. IP multicast (network-layer multicast) is implemented by network nodes (i.e., routers) and avoids multiple copies of the same datagram on the same link. Despite the conceptual simplicity of IP multicast and its obvious benefits, it has not been widely deployed since there remain many unresolved issues. As an alternative to IP multicast, overlay multicast (application-layer multicast) implements the multicast functionality at end hosts rather than routers. This may require more overall bandwidth than IP multicast because duplicate packets travel the same physical links multiple times, but it provides an inexpensive, deployable method of providing point-to-multipoint group communication. In this paper we develop an efficient method applied greedy algorithm for solving two models of overlay multicast tree building problem that is aimed to construct MDST (Minimum Diameter Spanning Tree : minimum cost path from a source node to all its receivers) and MST (Minimum Spanning Tree : minimum total cost spanning all the members). We also simulate and analyze MDST and MST.

• The KIPS Transactions:PartA
• /
• v.19A no.1
• /
• pp.1-8
• /
• 2012

Network-based intelligent robot is connected with network system, provides interactions with humans, and carries out its own roles on ubiquitous computing environments. URC (Ubiquitous Robot Companion) robot has been proposed to develop network-based robot by applying distributed computing techniques. On URC robot, it is possible to save the computing power of robot client by environments, has been proposed to develop robot software using service-oriented architecture on server-client computing environments. The SOMAR client robot consists of two layers - device service layer and robot service layer. The device service controls physical devices, and the robot service abstracts robot's services, which are newly defined and generated by combining many device services. RSEL (Robot Service Executing Language) is defined in this paper to represent relations and connections between device services and robot services. A RESL document, including robot services by combining several device services, is translated to a programming language for robot client system using RSEL translator, then the translated source program is compiled and uploaded to robot client system with RPC (Remote Procedure Call) command. A SOMAR client system is easy to be applied to embedded systems of host/target architecture. Moreover it is possible to produce a light-weight URC client robot by reducing workload of RSEL processing engine.

• The KIPS Transactions:PartC
• /
• v.8C no.6
• /
• pp.775-782
• /
• 2001

Recently the use of Linux OS is increasing to tremendous figures. But due to the fact that Linux is distributed on an open-source policy, the need of security is an upcoming question which leads to widespread development of security on a Linux based environment. Cryptography, however, can cause various problems because of difficulty of key management. A lot of researchers have been concentrating on the key recovery technique to eliminate the reverse effect of using these kinds of security and to promote positive aspects of using it. In this thesis I am suggesting an mechanism based on the key recovery technique, as a method to save time in recovery and resetting a disconnection between two end-users through IPSec (IP Security) protocols in a VPN (Virtual Private Network) environment. The main idea of the newly suggested mechanism, KRFSH (Key Recovery Field Storage Header), is to store the information of the session in advance for the case of losing the session information essential to establish a tunnel connection between a SG and a host in the VPN environment, and so if necessary to use the pre-stored information for recovery. This mechanism is loaded on the IPSec based FreeS/WAN program (Linux environment), and so the VPN problem mentioned above is resolved.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.1
• /
• pp.35-40
• /
• 2019

It is important to constantly provide service in real-time multimedia applications using multicast. Transmission path reconstruction occurs in hybrid multicast using Internet Protocol (IP) multicast and ALM in order to adapt the network status to things like congestion. So, there is a problem in which real-time QoS is reduced, caused by an increase in end-to-end delay. In this paper, we want to solve this problem through multi-layer transmission path construction. In the proposed method, we deploy the control server and application layer overlay host (ALOH) in each multicast domain (MD) for hybrid multicast construction. After the control server receives the control information from an ALOH that joins the MD, it makes a group based on the hop count and sends it to the ALOH in each MD. The ALOH in the MD performs the role of sending the packet to another ALOH and constructs the multi-layered transmission path in order of priority by using control information that is received from the control server and based on the delay between neighboring ALOHs. When congestion occurs in, or is absent from, the ALOH in the upper MD, the ALOH selects the path with the highest priority in order to reduce end-to-end delay. Simulation results show that the proposed method could reduce the end-to-end delay to less than 289 ms, on average, under congestion status.