• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.2
• /
• pp.310-323
• /
• 2004

In this paper, we extend the multiple time scale control framework to window-based congestion control, in particular, TCP This is performed by interfacing TCP with a large tine scale control nodule which adjusts the aggressiveness of bandwidth consumption behavior exhibited by TCP as a function of "large time scale" network state. i.e., conformation that exceeds the horizon of the feedback loop as determined by RTT Our contribution is threefold. First, we define a modular extension of TCP-a function call with a simple interface-that applies to various flavors of TCP-e.g., Tahoe, Reno, Vegas and show that it significantly improves performance. Second, we show that multiple time scale TCP endows the underlying feedback control with preactivity by bridging the uncertainty gap associated with reactive controls which is exacerbated by the high delay-bandwidth product in broadband wide area networks. Third, we investigate the influence of three traffic control dimensions-tracking ability, connection duration, and fairness-on performance. Performance evaluation of multiple time scale TCP is facilitated by a simulation bench-mark environment which is based on physical modeling of self-similar traffic.

• International Journal of Control, Automation, and Systems
• /
• v.3 no.1
• /
• pp.122-129
• /
• 2005

We describe the MiniWeb[7] TCP/IP stack (mIP), which is an extremely small implementation of the TCP/IP protocol suite running 8 or 32-bit micro controllers intended for embedded control systems, and satisfying the subset of RFC1122 requirements needed for host­to-host interoperability over different platforms. Our TCP/IP implementation does sacrifice some of TCP's mechanisms such as fragmentation, urgent data, retransmission, or congestion control. Our implementation is applicable to web based controllers. The network protocols are tested in operational networks using CommView and Dummynet where the various operational parameters such as bandwidth, delay, and queue sizes can be set and controlled.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06d
• /
• pp.430-432
• /
• 2012

최근 데이터 센터의 성장이 급속화됨에 따라, 네트워크 내부에서 전송되는 플로우들의 성능을 향상시키는 방법에 많은 관심이 생겨나고 있다. 특히 데이터 센터 내부의 스위치에서 생겨나는 큐잉 지연을 줄이는 것이 중요한 문제로 떠올랐다. 본 논문에서는 인터넷에서 널리 사용되고 있는 TCP 알고리즘이 데이터 센터에서 사용되었을 때 가지게 되는 근본적인 문제점을 파악하고, 이를 바탕으로 데이터 센터에서 낮은 지연 시간을 유지할 수 있는 지연 시간과 속도에 기반한 혼잡 제어 알고리즘을 제안한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.11
• /
• pp.5179-5202
• /
• 2018

Transmission Control Protocol (TCP) is the most widely used protocol in the cloud data centers today. However, cloud data centers using TCP experience many issues as TCP was designed based on the assumption that it would primarily be used in Wide Area Networks (WANs). One of the major issues with TCP in the cloud data centers is the Incast issue. This issue arises because of the many-to-one communication pattern that commonly exists in the modern cloud data centers. In many-to-one communication pattern, multiple senders simultaneously send data to a single receiver. This causes packet loss at the switch buffer which results in TCP throughput collapse that leads to high Flow Completion Time (FCT). Recently, Software-Defined Networking (SDN) has been used by many researchers to mitigate the Incast issue. In this paper, a detailed survey of various SDN based solutions to the Incast issue is carried out. In this survey, various SDN based solutions are classified into four categories i.e. TCP Receive Window based solutions, Tuning TCP Parameters based solutions, Quick Recovery based solutions and Application Layer based solutions. All the solutions are critically evaluated in terms of their principles, advantages, and shortcomings. Another important feature of this survey is to compare various SDN based solutions with respect to different performance metrics e.g. maximum number of concurrent senders supported, calculation of delay at the controller etc. These performance metrics are important for deployment of any SDN based solution in modern cloud data centers. In addition, future research directions are also discussed in this survey that can be explored to design and develop better SDN based solutions to the Incast issue.

• Journal of the Korean Society of Safety
• /
• v.18 no.3
• /
• pp.154-160
• /
• 2003

Uncertain time delay happens when the process reads the sensor data and sends the control input to the plant located at a remote site in distributed control system. As in the case of data network using TCP/IP, VDN that integrates both device network and data network has uncertain time delay. Uncertain time delay can cause degradation in performance and stability of distributed control system based on VDN. This paper first investigates the transmission characteristic of VDN and suggests a control scheme based on the Smith's predictor to minimize the effect of uncertain varying time delay. The validity of the proposed control scheme is demonstrated with real-time velocity control of DC servo motor located in remote site.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.9B
• /
• pp.829-837
• /
• 2006

Due to the internet network congestion, packets may be dropped or delayed at routers. This phenomenon degrades the quality of streaming applications that require high QoS requirements. The proposed algorithm in this paper, called DBRC(Delay-Based Rate Control), tries to cause router queue occupancy to reach a steady state or equilibrium by throttling the transmission rate of the multimedia traffics when network delays tend to increase and also probing for more bandwidth when network delays tend to decrease. Simulation results show that the proposed algorithm provides smooth transmission rate, nearly constant delay and low packet loss rates, compared with TFRC(TCP Friendly Rate Control) that is one of dominant multimedia congestion control algorithms.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.2 no.1
• /
• pp.10-18
• /
• 2007

In this paper, we extend the multiple time scale control framework to window-based congestion control, in particular, TCP. This is performed by interfacing TCP with a large time scale control module which adjusts the aggressiveness of bandwidth consumption behavior exhibited by TCP as a function of "large time scale" network state. i.e., conformation that exceeds the horizon of the feedback loop as determined by RTT. Performance evaluation of multiple time scale TCP is facilitated by a simulation bench-mark environment which is based on physical modeling of self-similar traffic. If source traffic is not extended exceeding, when RTT is 450ms, in self similar burst environment, performance gain of TCP-SSC is up to 45% for ${\alpha}$=1.05. However, its is acquired only 20% performance gain for ${\alpha}$=1.95 relatively. Therefore we showed that by TCP-MTS at large time scale into a rate-based feedback congestion control, we are able to improve two times performance significantly.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.8B
• /
• pp.699-706
• /
• 2008

This paper compares throughput performance of TCP and SCTP in a variety of network environments. For experiments, we construct a Linux-based testbed and consider a set of performance metrics such as MSS(Maximum Segment Size), transmission delay, and packet loss rate. In addition, we analyze the effect of SCTP multi-streaming on throughput. From the experimental results, we can see that SCTP provides throughput gain of approximately $20%{\sim}50%$ over TCP. This performance gain comes from the distinctive features of SCTP such as chunk bundling, initial congestion window of 2 MTU and SACK(Selective ACK) based error control. In the lossy networks, we can see that SCTP multi-streaming transmissions can effectively overcome the so-called HoLB(Head-of-Line Blocking) phenomenon of TCP.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.8
• /
• pp.1391-1400
• /
• 2008

In recent yeats, SCTP is known as the next transport layer protocol for connection-oriented and reliable data transfer after TCP/UDP. The SCTP was developed on the basis of the existing TCP and was designed to eliminate defects of TCP. SCTP has the different characteristic of multi-streaming and multi-homing from TCP. In this paper, we studied the defects of the performance of communication using the traits of multi-homing. And we evaluated the efficiency for each application services on the basis of SCTP of IPv6. Also, we analysed the delay time of a packet using SCTP and TCP. We noticed that the efficiency of SCTP is better than that of TCP.

• Journal of Communications and Networks
• /
• v.8 no.3
• /
• pp.351-359
• /
• 2006

In this article, we present a new slow-start variant, which improves the throughput of transmission control protocol (TCP) Vegas. We call this new mechanism Gallop-Vegas because it quickly ramps up to the available bandwidth and reduces the burstiness during the slow-start phase. TCP is known to send bursts of packets during its slow-start phase due to the fast window increase and the ACK-clock based transmission. This phenomenon causes TCP Vegas to change from slow-start phase to congestion-avoidance phase too early in the large bandwidth-delay product (BDP) links. Therefore, in Gallop-Vegas, we increase the congestion window size with a rate between exponential growth and linear growth during slow-start phase. Our analysis, simulation results, and measurements on the Internet show that Gallop-Vegas significantly improves the performance of a connection, especially during the slow-start phase. Furthermore, it is implementation feasible because only sending part needs to be modified.