• 대한공업교육학회지
• /
• v.33 no.2
• /
• pp.248-258
• /
• 2008

Current Internet uses HTTP (Hyper Text Transfer Protocol) as an application layer protocol and TCP (Transmission Control Protocol) as a transport layer protocol to provide web service. SCTP (Stream Control Transmission Protocol) is a recently proposed transport protocol with very similar congestion control mechanisms as TCP, except the initial congestion window during the slow start phase. In this paper, we present a mathematical model of object transfer latency during the slow start phase for HTTP over SCTP and compare with the latency of HTTP over TCP. Validation of the model using experimental result shows that the mean object transfer latency for HTTP over SCTP during the slow start phase is less than that for HTTP over TCP by 11%.

• The KIPS Transactions:PartC
• /
• v.11C no.7 s.96
• /
• pp.931-936
• /
• 2004

We need to have an adaptive TCP protocol that can be tolerable on wireless network environement. TCP Westwood for use in the environe-ment that have a very high loss rate like a sattelite was proposed by modifying the existing bulk retransmission protocol. Bulk retransmission mechanism shows a highly enhanced performance on networks that have a very high loss rate but are prone to bursty loss networks. Also, it can exprience less performance on low late transmission environement. This paper proposes Adaptive Bulk Retransmission Mechanism that adjusts the number of bulk retransmitted packets based on the network conditions. The proposed mechanism was evaluated by using NS-2.

• Proceedings of the KIEE Conference
• /
• 2008.10b
• /
• pp.544-545
• /
• 2008

무선 통신 기술의 발달로 최근 무선 device 들로 이루어진 ad-hoc network이 주요 issue가 되고 있으며, ad-hoc network에서의 성능이 중요한 관심사가 되고 있다. 그러나 유선망에 최적화되도록 설계된 TCP 는 무선망이 가지는 링크의 불안정함으로 인한 손실을 네트워크 혼잡에 의한 손실로 인식하여 혼잡 제어 메커니즘이 수행함으로써 전송률을 줄이게 되므로 전체적으로 네크웍의 성능이 저하되는 문제점을 가진다. 본 논문에서는 무선망에서 TCP 성능을 향상시키기 위한 새로운 congestion algorithm을 제안하고 그 성능을 평가하였다. 성능 확인 결과 기존 TCP-vegas에 비해 더 나은 성능을 보임을 알 수 있다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06d
• /
• pp.137-140
• /
• 2008

TCP는 유선망에 최적화되도록 변화 된 결과, 무선망에서는 무선망의 단점인 비연속적인 연결 링크로 인해 전송손실이 발생하게 된다. 그러나 어플리케이션 프로그램에서는 이러한 문제로 인해 발생되는 오류를 네트워크의 혼잡으로 인한 손실로 오인하게 된다. 그 결과 어플리케이션 프로그램에서는 혼잡제어 메커니즘이 수행되어 전송율을 줄이므로 네트워크 성능이 저하되는 문제점을 초래한다. 이러한 이유로 최근 유무선 혼합망에서 TCP의 성능을 향상시키기 위한 연구가 활발히 이루어지고 있다. 본 논문에서는 기존에 제안된 개선 모델들 중 상대적으로 많이 사용되고 있는 TCP Westwood에 대해서 성능을 보완하는 기법을 제안하고자 한다.

• Journal of Broadcast Engineering
• /
• v.10 no.3
• /
• pp.297-312
• /
• 2005

The multimedia traffic of continuous video and audio data via streaming service accounts for a significant and expanding portion of the Internet traffic. This streaming data delivery is mostly based on RTP with UDP. However, UDP does not support congestion control. For this reason, UDP causes the starvation of congestion controlled TCP traffic which reduces its bandwidth share during overload situation. In this paper, we propose a new TCP-friendly rate control scheme called 'TF-RTP(TCP-Friendly RTP)'. In the congested network state, the TF-RTP exactly estimates the competing TCP's throughput by using the modified parameters. Then, it controls the sending rate of the video streams. Therefore, the TF-RTP adjusts its sending rate to TCP-friendly and fair share with competing TCP traffics. Through the simulation, we prove that the TF-RTP correctly estimates the TCP's throughput and improves the TCP-friendliness and fairness.

• Journal of KIISE:Computer Systems and Theory
• /
• v.34 no.9
• /
• pp.425-434
• /
• 2007

Even though Internet backbone speeds have increased in the last few years due to projects like Internet 2 and NGI, many high performance distributed applications are able to achieve only a small fraction of the available bandwidth. The cause of such problem is due to a character of TCP/IP. The primary goal of this protocol is reliable data transmission. Therefore high speed data transmission didn't be considered when TCP/IP is designed. Hence several researchers have been studied in order to solve the problem of TCP/IP. One of these research results, parallel transfer technique, solves this problem to use parallel TCP connections on application level. Additionally, this technique is compatibility. Recently, these researchers have been studied a mechanism to decide the number of parallel TCP connections. However, some researchers reported the number of parallel TCP connection base on only empirical results. Although hardware performance of host affects transmission rate, the hardware performance didn't be considered in their works. Hence, we collect all data related to transmission rate, such as hardware state information (cpu utilization, interrupt, context switch). Then, we analyzed collected data. And, we suggest a new mechanism determining number of parallel TCP connections for maximization of performance based on our analysis.

• The KIPS Transactions:PartC
• /
• v.9C no.2
• /
• pp.197-212
• /
• 2002

Transmission Control Protocol (TCP) [1] has been tuned as a reliable transfer protocol for traditional networks comprising wired links and stationary hosts with same link characteristics. TCP assumes that congestion in the network be a primary cause for packet losses and unusual delays. TCP performs welt over such networks adapting to end-to-end delays and congestion losses, by standard congestion control mechanisms, such as slow-start, congestion avoidance, fast retransmit and recovery. However, networks with wireless and other lossy links suffer from significant losses due to bit errors and handoffs. An asymmetry network such as ADSL has different bandwidth for both directions. As a result, TCP's standard mechanisms incur end-to-end performance degradation in various links. In this paper, we analyze the TCP problems in wireless, satellite, and asymmetry links, and measure the new TCP mechanisms that are recommended by IETF Performance Implications of Link Characteristics (PILC) WG[2], by using Network Simulator 2 (NS-2).

• The KIPS Transactions:PartC
• /
• v.11C no.2
• /
• pp.245-252
• /
• 2004

Stream Control Transmission Protocol(SCTP) is a new connection-oriented, reliable delivery transport protocol operating on top of an unreliable connectionless packet service such as IP. It inherits many of the functions developed for TCP, including flow control and packet loss recovery functions. In addition, it also supports transport layer multihoming and multistreaming In this paper, we study the impact of multi-homing on the performance of SCTP. We first compare performance of single-homed SCTP. multi-homed SCTP, TCP Reno and TCP SACK. We, then describe potential flaw in the current SCTP retransmission policy, when SCTP host is multihomed. Our Results show that SCTP performs better than TCP Reno and TCP SACK due to several changes from TCP in its congestion control mechanism. In particular. multi-homed SCTP shows the best result among the compared schemes. Through experimentation for multi-homed SCTP, we found that the current SCTP retransmission policy nay deteriorate the perfomance when the retransmission path it worse than the original path. Therefore, the condition of retransmission path is a very important factor In SCTP performance and a proper mechanism would be required to measure the condition of the retransmission path.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04a
• /
• pp.16-18
• /
• 2003

클러스터 파일 시스템은 데이터 입출력 대역폭을 극대화하여 효율성을 높이고 각 노드의 입출력 부담을 균등하게 부과하기 위하여 원본 파일을 여러 노드에 분산 저장한다. 이렇게 파일을 노드들에 분산 저장하기 위해서는 효율적인 노드간 데이터 통신을 필요로 하며, 노드 내부에서도 클러스터 파일 시스템과 어플리케이션과의 효율적인 전용 데이터 교환 메커니즘을 지원해야 한다. 본 논문에서는 WAN(Wide Area Network)에 적합하도록 설계된 TCP를 이용한 기존의 교환 메커니즘인 통신모듈이 가지고 있는 문제점을 해결하기 위해 다양한 프로토콜과 하드웨어적인 접근을 통해 Myrinet이 초고속 통신모듈에 적합함을 보이고 GM API를 활용하여 기존의 소켓기반인 TCP/IP를 이용한 통신 모듈을 대처할 새로운 통신모듈의 모델을 제시한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2005.05a
• /
• pp.1465-1468
• /
• 2005

유선망에 최적화 되도록 설계된 TCP는 무선망에 적용되었을 경우 무선망의 특성에 의해 발생하는 데이터의 손실을 망의 혼잡에 의한 손실로 인식하여 불필요한 혼잡 제어 메커니즘을 수행하여 망의 성능을 떨어뜨리는 문제가 발생 된다. 본 논문에서는 이 같은 문제점을 보완하기 위해 제안된 Snoop의 성능을 TCP와 비교 분석하고 앞으로 진행될 연구를 제안한다.