• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.8A
• /
• pp.1096-1103
• /
• 2000

The default server strategy is commonly used for location and state managements of mobile host in mobile computing. With this strategy, we can find the cell of destination mobile host to send data by querying the default server. In SDN(single Default Notification) strategy which is a kind of default server strategy, the call is established after the location and state of the callee is acquired to the query server by querying the default server. But the communication cost overhead from the default server is increased if there are large number of cells and query requests, and if it is too far from the default server to a base station. Still more it will be unable to establish any calls to a mobile host when there is a fault in the default server of this host. In this paper, we suggest add evaluate a default server replication strategy to reduce the communication cost overhead and to make the service available.

• Journal of KIISE:Information Networking
• /
• v.29 no.3
• /
• pp.265-275
• /
• 2002

A default server strategy is a very popular scheme for managing location and state information of mobile hosts in cellular networks. But the communication cost increases if the call requests are frequent and the distant between the default server and the client is long. Still more any connection to a mobile host cannot be established when the default server of the destination mobile host fails. These problems can be solved by replicating default server and by letting nearest replicated default server process the query request which is sent from a client. It is important to allocate replicated default servers efficiently in networks and determine the number of replicated default servers. In this paper, we suggest and evaluate a default server replication strategy to reduce communication costs and to improve service availabilities. Furthermore we propose and evaluate an optimized allocation algorithm and an optimal replication degree for replicating: dofault servers in nn grid networks and binary tree networks.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.04a
• /
• pp.507-510
• /
• 2001

HTTP/1.1 standard reduces latencies and overhead from closing and re-establishing connections by supporting persistent connections as a default, which encourage multiple transfers of objects over one connection. HTTP/1.1, however, does not define explicitly connection-closing time but specifies a certain fixed holding time model. This model may induce wasting server's resource when server maintains conn ection with the idle-state client that requests no data for a certain time. This paper proposes the mechanism of a heuristic connection management supported by the client-side under persistent HTTP, in addition to HTTP/1.1's fixed holding time model on server-side. The client exploits the tag information within transferred HTML page so that decides connection-closing time. As a result, the mechanism allows server to use server's resource more efficiently without server's efforts.

• Journal of IKEEE
• /
• v.7 no.2 s.13
• /
• pp.265-270
• /
• 2003

We divide networks of an organization into internal network and external network to distribute the load of web server, and allocate separate web server for each network. The performance of web server for internal networks can be affected seriously by RTO(Retransmission Timeout). The value of RTO set by default in the OS of an web server is so large that it degrades the performance of web server. Therefore, this paper suggests not useing but applying conventional algorithm to calculate RTO, but to apply newly controlled value of RTO, and it showed improvement of the performance of web server.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.04a
• /
• pp.355-357
• /
• 2001

HTTP/1.1 standard reduces latencies and overhead from closing and re-establishing connections by supporting persistent connections as a default, which encourage multiple transfers of objects over one connection. HTTP/1.1, however, does not define explicitly connection-closing time but specifies a certain fixed holding time model. This model may induce wasting server’s resource when server maintains connection with the idle-state client that requests no data for a certain time. This paper proposes the mechanism of a heuristic connection management supported by the client-side under persistent HTTP, in addition to HTTP/1.1’s fixed holding time model on server-side. The client exploits the tag information within transferred HTML page so that decides connection-closing time. As a result, the mechanism allows server to use server’s resource more efficiently without server’s efforts.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10c
• /
• pp.250-252
• /
• 1999

모빌 컴퓨팅에서는 모빌 호스트의 위치와 상태관리를 위하여 디폴트 서버기법이 널리 사용되는데, 모빌 호스트로 전송되는 데이터는 먼저 디폴트 서버로 보내진 후 다시 모빌 호스트가 위치하는 로칼 서버로 전송된다. 디폴트 서버기법에 기반을 둔 SDN(single Default Notification)기법에서는 클라이언트가 모빌 호스트와의 연결요청을 하면 쿼리 서버를 통해 해당 디폴트 서버에 모빌 호스트의 위치 및 상태를 문의한 후 통신이 이뤄진다. 그러나 쿼리 횟수가 많고 디폴트 서버와의 거리가 멀거나 기지국의 수가 많을 경우 디폴트 서버와의 통신 오버헤드가 커지며, 디폴트 서버에 결함이 발생할 경우 해당 모델 호스트와의 연결이 불가능해진다. 본 논문에서는 위와 같은 문제점을 해결하기 위해 디폴트 서버와의 통신 비용을 줄이기 위한 디폴트 서버 중첩 기법을 제안한다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.2
• /
• pp.571-574
• /
• 2005

XCAP(XML Configuration Access Protocol) which has been proposed in IETF is based on both XML and HTTP protocol. XCAP server maintains user's configuration information for specific application which is described by XML. This protocol can be applied to many application servers for adapting user's preferences. There can be many way to interwork with other application servers. In this paper, we will talk about the experience of designing and implementation of XCAP server and the way of interwork with application servers.

• Journal of Information Processing Systems
• /
• v.19 no.2
• /
• pp.240-257
• /
• 2023

The industrial Internet of Things (IIoT) is characterized by intelligent connection, real-time data processing, collaborative monitoring, and automatic information processing. The heterogeneous IIoT devices require a high data rate, high reliability, high coverage, and low delay, thus posing a significant challenge to information security. High-performance edge and cloud servers are a good backup solution for IIoT devices with limited capabilities. However, privacy leakage and network attack cases may occur in heterogeneous IIoT environments. Cloud-based multi-party computing is a reliable privacy-protecting technology that encourages multiparty participation in joint computing without privacy disclosure. However, the default cloud selection method does not meet the heterogeneous IIoT requirements. The server can be dishonest, significantly increasing the probability of multi-party computation failure or inefficiency. This paper proposes a blockchain and smart contract-based optimized cloud node selection framework. Different participants choose the best server that meets their performance demands, considering the communication delay. Smart contracts provide a progressive request mechanism to increase participation. The simulation results show that our framework improves overall multi-party computing efficiency by up to 44.73%.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.16 no.5
• /
• pp.201-207
• /
• 2021

Edge Computing enables image-based inference in close proximity to end users and real-world objects. However, since edge servers have limited computational and accelerator resources, efficient resource management is essential. In this paper, we present HeteroAccel system that performs optimal scheduling in Kubernetes platform based on available node and accelerator information for various inference requests. Our experiments showed 25.3% improvement in overall inference performance over the default scheduling scheme in edge computing environment in which four types of inference services are requested.

• KIISE Transactions on Computing Practices
• /
• v.24 no.1
• /
• pp.10-16
• /
• 2018

The task of consolidating server workloads is critical for the efficiency of a datacenter in terms of reducing costs. However, as a greater number of workloads are consolidated in a single server, the performance of workloads might be degraded due to their contention to the limited shared resources. To reduce the performance degradation, scheduling for mitigating the contention of shared resources is necessary. In this paper, we present the Dynamic Voltage Frequency Scaling (DVFS) based memory-contention aware scheduling method for multi-threaded workloads. The proposed method uses two approaches: running memory-intensive threads on the limited cores to avoid concurrent memory accesses, and reducing the frequencies of the cores that run memory-intensive threads. With the proposed algorithm, we increased performance by 43% and reduced power consumption by 38% compared to the Completely Fair Scheduler(CFS), the default scheduler of Linux.