• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.6 no.2
• /
• pp.236-247
• /
• 2002

We propose a cell－multiplexing scheme for the real－time communication service in ATM network and a new service discipline guarantee end-to-end delay based on pseudo-isochronous cell switching. The proposed scheme consists of two level frame hierarchy, upper and lower frame, which is used to assign the bandwidth and to guarantee the requested delay bound, respectively. Since the Proposed algorithm employs two level frame hierarchy, it can overcome the coupling problem which is inherent to the framing strategy It can be comparatively reduce the complexity, and still guarantee the diverse delay bounds of end-to-end. Besides, it consists of two components, traffic controller and scheduller, as the imput traffic description model and regulates the input traffic specification. The function of the traffic controller is to shape real -time traffic to have the same input pattern at every switch along the path. The end-to-end delay is bounded by the scheduller which can limit the delay variation without using per-session jitter controllers, and therefore it can decrease the required buffer size. The proposed algorithm can support the QoS's of non-real time traffic as well as those of real time traffic.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.5C
• /
• pp.445-453
• /
• 2002

We propose a cell-multiplexing scheme for the real-time communication service in ATM network and a new service discipline guarantee end-to-end delay based on pseudo-isochronous cell switching. The proposed scheme consists of two level frame hierarchy, upper and lower frame, which is used to assign the bandwidth and to guarantee the requested delay bound, respectively. Since the proposed algorithm employs two level frame hierarchy, it can overcome the coupling problem which is inherent to the framing strategy. The proposed scheme consists of two components, traffic controller and scheduller, as the imput traffic description model and regulates the input traffic specification. The function of the traffic controller is to shape real-time traffic to have the same input pattern at every switch along the path. The end-to-end delay is bounded by the scheduller which can limit the delay variation without using per-session jitter controllers, and therefore it can decrease the required buffer size. The proposed algorithm can support the QoS's of non-real time traffic as well as those of real time traffic

• The KIPS Transactions:PartC
• /
• v.10C no.1
• /
• pp.45-50
• /
• 2003

Guaranteed Frame Rate (GFR) service category is one of the most recent ATM service categories. The GFR specification was recently finalized by the ATM Forum and is expected to become an important service category which can efficiently support TCP/IP traffic in ATM network. In GFR switch implementation, it is important to guarantee MCR (minimum cell rate) and improve fairness. In this paper, we propose a new scheduling algorithm for the GFR service. Proposed algorithm provides priority to VC (virtual circuit)s and high Priority given to a VC which has fewer untagged cells in buffer. High priority VCs are serviced before low priority VCs. Proposed algorithm can guarantee MCR and improve fair sharing of the remaining bandwidth between GFR VCs. From computer simulation results, we demonstrate the proposed scheduling algorithm provide much better performance in TCP goodput and fairness than previous schemes.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.12B
• /
• pp.838-846
• /
• 2005

In this paper, we consider the following two issues for implementing the synchronous Ethernet systems. First, a synchronous Ethernet system employs a fixed size superframe which is divided into a synchronous period and an asynchronous one. We note that the starting point of a superframe is not deterministic when an ordinary data frame's transmission is overlapped the superframe boundary. This overlap may be a fatal drawback for strict jitter bounded applications. Circumventing the problem, we propose a frame fragmentation scheme to provide a zero jitter, and compare its delay performance with the hold scheme which also provides the zero jitter. We next concern that IEEE 802.3x pause frames cannot be promptly transmitted in a synchronous period, and thus asynchronous traffics may be dramatically get dropped at the input buffer of a switch. To handle the problem, we propose an efficient flow control by allowing the transmission of the pause frame in a synchronous period, and investigate the blocking probability of the asynchronous traffics by the simulation.

• Journal of Sensor Science and Technology
• /
• v.7 no.4
• /
• pp.293-299
• /
• 1998

An objective of this study is to develop a voltage measuring device that uses a gas-filled switch (GS) on 22.9 kV-y extra-high voltage distribution lines. The voltage measuring device proposed in this paper is a kind of capacitive divider which consists of a detecting electrode attached outside of the bushing of GS, an impedance matching circuit, and a voltage buffer. It can be easily installed in an established GS without changing the structure. For the calibration and application investigations, the voltage measuring device was set up in the 25.8 kV 400 A GS, and a step pulse generator having 5 ns rise time is used. As a result, it was found that the frequency bandwidth of the voltage measuring device ranges from 1.35 Hz to about 13 MHz. The error of voltage dividing ratio which is evaluated by the commercial frequency voltage of 60 Hz was less than 0.2%. In addition, voltage dividing ratio in the commercial frequency voltage and in a non-oscillating impulse voltage were compared, and their deviation were less than 0.7%.

• 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 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.