This paper gives two reasons for incast, namely, block tail time out (BTTO) and block head time out (BHTO). The former is about the lost of one of the several last packet, which cannot produce enough dupack to signal fast retransmit; the latter is about the lost of whole window...
[more]
Zhang et al (2011) Shrinking MTU to Mitigate TCP Incast Throughput Collapse in Data Center Networks (ICCMC)
A short paper that proposes to use a smaller MTU to solve the incast problem. As the buffer size is fixed, a smaller packet means the buffer can hold more packets, and then it allows the TCP window to be “larger”. A larger TCP window is easier to catch packet...
[more]
Shpiner & Keslassy (2010) A Switch-Based Approach to Throughput Collapse and Starvation in Data Centers (IWQoS)
Solution to incast problem using switch, transparent to the end hosts.
[more]
Wu et al (2010) ICTCP: Incast Congestion Control for TCP in Data Center Networks (CoNEXT)
A paper to address the incast issue in data center TCP.
[more]
Devkota & Reddy (2010) Performance of Quantized Congestion Notification in TCP Incast Scenarios of Data Centers (MASCOTS)
This paper studies the problem of TCP in-cast in 802.1Qau network environment. This paper reveals that,
[more]
