Browsing by Author "Tsao, Shih-Chiang"

Now showing 1 - 6 of 6

A Fast-Converging TCP-Equivalent Window-Averaging Rate Control Scheme
(2012)
Tsao, Shih-Chiang
;
Lai, Yuan-Cheng
;
Lin, Ying-Dar
;
資訊工程學系
;
Department of Computer Science
Smooth rate control schemes are necessary for Internet streaming flows to use available bandwidth. To equally share the Internet bandwidth with existing Transmission Control Protocol (TCP) flows, these new schemes should meet TCP-equivalent criterion, i.e., achieve the same transmission rates as TCP under the same network conditions. However, when the available bandwidth oscillates, many of these schemes fail to meet this criterion due to their slow increasing rate. This study proposes a window-averaging rate control (WARC) scheme to send packets at the same average rate as a TCP flow over a fixed time interval. Considering the TCP rate only over a fixed interval allows WARC to forget the historical loss condition more rapidly than other schemes, thus achieving a faster increasing rate when additional bandwidth becomes available. When the available bandwidth drops dramatically, WARC uses a history-reset procedure to converge its rate to the new steady rate immediately after a specified number of losses. The simulations in this study show that WARC not only achieves the same bandwidth as TCP, but exhibits faster convergent behaviors and has a smoother rate than existing schemes.
Multiple-resource request scheduling for differentiated QoS at website gateway
(2008-06-25)
Lin, Ying-Dar
;
Tien, Ching-Ming
;
Tsao, Shih-Chiang
;
Feng, Ruo-Hua
;
Lai, Yuan-Cheng
Differentiated quality of service is a way for a website operator to provide different service levels to its clients. Traditional HTTP request scheduling schemes can achieve this, but they schedule requests to manage only one server resource, such as CPU or disk I/O. Actually, processing a request on the server will consume multiple resources. This paper presents a multiple-resource request scheduling algorithm, called mQoS, for differentiating the utilization of the server resource. The mQoS scheduler consists of several sub-schedulers and a main scheduler. Each sub-scheduler manages a server resource to differentiate its utilization among the classes. The main scheduler checks the availability of every server resource and triggers an appropriate sub-scheduler to balance the utilization of server resources. The implementation of the mQoS gateway is based on Squid and Linux. The evaluation compares the mQoS scheduling with no scheduling (nQoS) and single-resource request scheduling (sQoS). The mQoS scheduling reveals the accurate differentiation oil every server resource. In addition, the total server throughput in the mQoS scheduling is improved by 21%, compared with the sQoS scheduling. The average user-perceived latency of the mQoS scheduling is also shorter than other scheduling. (C) 2008 Elsevier B.V. All rights reserved.
On applying fair queuing discipline to schedule requests at access gateway for downlink differential QoS
(2008-12-22)
Tsao, Shih-Chiang
;
Lai, Yuan-Cheng
;
Tsao, Le-Chi
;
Lin, Ying-Dar
;
資訊工程學系
Scheduling packets is a usual solution to allocate the bandwidth on a bottleneck link. However, this solution cannot be used to manage the downlink bandwidth at the user-side access gateway, since the traffic is queued at the ISP-side gateway but not the user-side gateway. An idea is scheduling the requests at the user-side gateway to control the amount of the responses queued in the ISP-side gateway. This work first investigates the possibility of applying the class-based fair queuing discipline, which was widely and maturely used in scheduling packets, to schedule requests. However, we found that simply applying this discipline to schedule requests would encounter the timing and ordering problems at releasing requests and may not satisfy high-class users. Thus, we propose a minimum-service first request scheduling (MSF-RS) scheme. MSF-RS always selects the next request from the class receiving the minimum service to provide user-based weighted fairness, which ensures more bandwidth for high-class users. Next, MSF-RS uses a window-based rate control on releasing requests to maintain full link utilization and reduce the user-perceived latency. The results of analysis, simulation and field trial demonstrate that MSF-RS provides fairness while reducing 23-30% of user-perceived latency on average. Besides, a MSF-RS gateway can save 25% of CPU loading. (C) 2008 Elsevier B.V. All rights reserved.
On-the-fly TCP path selection algorithm in access link load balancing
(2007-01-15)
Lin, Ying-Dar
;
Tsao, Shih-Chiang
;
Leong, Un-Pio
;
資訊工程學系
;
Department of Computer Science
Many enterprises install multiple access links for fault tolerance or bandwidth enlargement. Dispatching connections through good links is the ultimate goal in utilizing multiple access links. The traditional dispatching method is only based on the condition of the access links to ISPs. It may achieve fair utilization on the access links but poor performance on connection throughput. In this work, we propose a novel approach to maximize the per-connection end-to-end throughput by the on-the-fly round trip time (RTT) probing mechanism. The end-to-end RTTs through all possible links are probed by duplicating the SYN packet during the three-way handshaking stage of a TCP connection. The experiment results show that the ratio to choose the best outgoing access link is 79% on the average. If the second best link is chosen, it is usually very close to the best, thus averagely achieving 94% of the maximum possible throughput. The ratio of the traditional round-robin (RR) algorithm is only 35%, and the link selected by RR algorithm could provide 69% of throughput. (c) 2006 Elsevier B.V. All rights reserved.
Request Scheduling for Differentiated QoS at Website Gateway
(2008)
Lin, Ying-Dar
;
Tien, Ching-Ming
;
Tsao, Shih-Chiang
;
Wen, Shuo-Yen
;
Lai, Yuan-Cheng
With the explosive growth of Web traffic, the load on a Web server becomes heavier, leading to the longer user-perceived latency. Website operators would like to employ service differentiation to offer better throughput and shorter user-perceived latency to some specific users. This paper presents an HTTP request scheduling algorithm deployed at the Website gateway to enable the Web quality of service without any modification to client or server software. A variation of the deficit round robin algorithm for packet scheduling and a window control mechanism are presented to decide the order and the releasing time of requests, respectively. The order is decided by the response size of the requests and the pre-defined service weights. The ratio of the service rate got by the service classes is determined by the weights, whereas the releasing time is decided by the service rate of the Web server. The evaluation shows the scheduling algorithm can provide service differentiation and improve server throughput and user-perceived latency. When the weight ratio 6:3:1 is assigned to three service classes, the QoS Website gateway makes them get 60%, 30%, and 10% of the overall throughput as expected, regardless whatever page sizes. In addition, the throughput and the user-perceived latency of the class with the largest weight can be improved by up to 176% and 69% of the QoS-disabled values, respectively.
Taxonomy and evaluation of TCP-Friendly congestion-control schemes on fairness, aggressiveness, and responsiveness
(2007-11-01)
Tsao, Shih-Chiang
;
Lai, Yuan-Cheng
;
Lin, Ying-Dar
;
交大名義發表
;
National Chiao Tung University
Many TCP-friendly congestion control schemes have been proposed to pursue the TCP-equivalence criterion, which states that a TCP-equivalent flow should have the some throughput with TCP if it experiences identical network conditions as TCP. Additionally, the throughput should converge as fast as TCP when the packet-loss conditions change. This study classifies eight typical TCP-friendly schemes according to their underlying policies on fairness, aggressiveness, and responsiveness. The schemes are evaluated to verify whether they meet TCP-equivalence and TCP-equal share. TCP-equal share is a more realistic but more challenging criterion than TCP-equivalence and states that a flow should have the some throughput with TCP if competing with TCP for the some bottleneck. Simulation results indicate that one of the selected schemes, TCP-friendly rate control (TFRC), meets both criteria under more testing scenarios than the others. Additionally, the results under nonperiodic losses, low-multiplexing, two-state losses, and bursty losses reveal the causes that bring fault cases to the schemes. Finally, appropriate policies are recommended for an ideal scheme.