Improving Perceived Web Performance by Size Based Congestion Control

Ziegler, Thomas; Tran, Hung Tuan; Hasenleithner, Eduard

doi:10.1007/978-3-540-24693-0_57

Cited by 5 publications

(2 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…al. [24] also proposed to dynamically adapt AIMD parameters for improving the startup latency of short flows. However, with their protocol, a short flow achieves no higher throughput than long flows for the protocol to be incentive-compatible with TCP.…”

Section: Related Workmentioning

confidence: 99%

Minimizing flow completion times in data centers

Munir

Qazi

Uzmi

et al. 2013

2013 Proceedings IEEE INFOCOM

182

View full text Add to dashboard Cite

For provisioning large-scale online applications such as web search, social networks and advertisement systems, data centers face extreme challenges in providing low latency for short flows (that result from end-user actions) and high throughput for background flows (that are needed to maintain data consistency and structure across massively distributed systems). We propose L 2 DCT, a practical data center transport protocol that targets a reduction in flow completion times for short flows by approximating the Least Attained Service (LAS) scheduling discipline, without requiring any changes in application software or router hardware, and without adversely affecting the long flows. L 2 DCT can co-exist with TCP and works by adapting flow rates to the extent of network congestion inferred via Explicit Congestion Notification (ECN) marking, a feature widely supported by the installed router base. Though L 2 DCT is deadline unaware, our results indicate that, for typical data center traffic patterns and deadlines and over a wide range of traffic load, its deadline miss rate is consistently smaller compared to existing deadlinedriven data center transport protocols. L 2 DCT reduces the mean flow completion time by up to 50% over DCTCP and by up to 95% over TCP. In addition, it reduces the completion for 99th percentile flows by 37% over DCTCP. We present the design and analysis of L 2 DCT, evaluate its performance, and discuss an implementation built upon standard Linux protocol stack.

show abstract

Section: Related Workmentioning

confidence: 99%

Minimizing flow completion times in data centers

Munir

Qazi

Uzmi

et al. 2013

2013 Proceedings IEEE INFOCOM

182

View full text Add to dashboard Cite

show abstract

“…Based on flow-sizes, authors of [9] and [10] have proposed new versions of TCP called TCP Vienna and TCP/SPAND respectively. The motivation for TCP Vienna is to minimize the unfairness against the small flows; and this can be achieved by increasing their throughput.…”

Section: Introductionmentioning

confidence: 99%

Development and experimentation of TCP initial window function

Barik

Divakaran

2014

2014 International Conference and Workshop on the Network of the Future (NOF)

View full text Add to dashboard Cite

In the light of the mice-elephant phenomenon of the Internet traffic, TCP congestion control algorithm shows unfairness against flows of small sizes (small flows). In this context, we find it motivating to study the influence of TCP's initial window (IW) size on the response times of small flows. In our previous works, we proposed a function that determines IW-size for each flow based on its size; through game-theoretic analysis as well as experiments on real testbed, we observed that the size-based IW function (in comparison to a single constant IW size) shows considerable improvement in the performance of small flows, while at the same time not affecting the performance of large flows. In this paper, we share the experience of using some important open source tools for developing, experimenting and evaluating the proposed IW function. We implement the size-based IW function in the Linux kernel, version 3.7.4; we use ipfw/Dummynet for emulating links, open source tool called parallel for executing the application in parallel for generating TCP traffic, wireshark (along with lua script) for extracting metrics for performance analysis. Finally we evaluate the performance of flows using the IW function.

show abstract