2018
DOI: 10.1109/jsyst.2016.2582527
|View full text |Cite
|
Sign up to set email alerts
|

Fast Coupled Retransmission for Multipath TCP in Data Center Networks

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
14
0

Year Published

2018
2018
2020
2020

Publication Types

Select...
5
1
1

Relationship

0
7

Authors

Journals

citations
Cited by 17 publications
(14 citation statements)
references
References 9 publications
0
14
0
Order By: Relevance
“…When a failure occurred on Path B from the 30th s, it caused back-to-back timeouts on Path B (see Figure 4; the cwnd size on Path B was always set to one after 30 s). As in the failure case, the MPTCP sender tried to transmit data after each of these timeouts on Path B (namely, unnecessary retransmission via the failure path [27]). The resulting receiver buffer blocking in MPTCP, with a great number of out-of-order data (see Figure 5) held in the constrained receiver buffer, prevented the efficient use of Path A for data transmission and resulted in serious throughput performance degradations.…”
Section: Problem Statementmentioning
confidence: 99%
“…When a failure occurred on Path B from the 30th s, it caused back-to-back timeouts on Path B (see Figure 4; the cwnd size on Path B was always set to one after 30 s). As in the failure case, the MPTCP sender tried to transmit data after each of these timeouts on Path B (namely, unnecessary retransmission via the failure path [27]). The resulting receiver buffer blocking in MPTCP, with a great number of out-of-order data (see Figure 5) held in the constrained receiver buffer, prevented the efficient use of Path A for data transmission and resulted in serious throughput performance degradations.…”
Section: Problem Statementmentioning
confidence: 99%
“…However, The regular MPTCP usually does not handle short flows in data center networks effectively because of the long delay from excessive timeouts. In [24] and [18], two methods were presented to overcome this limitation in standard MPTCP. The first one proposed Maximum MultiPath TCP (MMPTCP), which operates in two phases.…”
Section: Related Workmentioning
confidence: 99%
“…MPTCP utilizes multiple interfaces in two communicating peers to establish simultaneous sub-connection paths between the two. It can utilize full bisection bandwidth in topologies like dual-homed FatTree and BCube by creating multiple subflows between two servers [18]. In Amazon's EC2 data centers, MPTCP with four Figure 2: SDN network compared to traditional network [30] subflows achieves three times the throughput of a single-path TCP (SPTCP) [32].…”
mentioning
confidence: 99%
“…Nowadays, there are a lot of researchers paying attention to retransmission mechanism. Hwang et al [10] designed a fast coupled retransmission mechanism for MPTCP, call TastCoRe, to quickly retransmit data through the other non-congested available path. Shen et al [11] presented an improved retransmission policy (RTX-CR), which chooses the retransmission path by considering congestion window and RTT for reducing rbuf blocking.…”
Section: Background and Motivationmentioning
confidence: 99%
“…Unfortunately, the regular MPTCP employs a conservative retransmission mechanism, which retransmits packets in the same path, to transfer retransmission packets. Only when retransmission timeout, it selects another path to transfer resulting in the failure of a packet arriving at the receiver timely [10]. Furthermore, multiple retransmission timeout will lead to out‐of‐order packets arrival problem.…”
Section: Introductionmentioning
confidence: 99%