TCP incast is a recently identified network transport pathology that affects many-to-one communication patterns in datacenters. It is caused by a complex interplay between datacenter applications, the underlying switches, network topology, and TCP, which was originally designed for wide area networks. We seek to understand how incast impacts the emerging class of big data workloads. We develop and validate a quantitative model that accurately predicts the onset of incast and TCP behavior both before and after. We also investigate how incast affects the Apache Hadoop implementation of MapReduce, an important example of a big data application. We further reflect on some technology and data analysis trends surrounding big data, speculate on how these trends interact with incast, and make recommendations for datacenter operators.
Understanding TCP Incast and Its Implications for Big Data Workloads
Researchers may make free and open use of the UC Berkeley Library’s digitized public domain materials. However, some materials in our online collections may be protected by U.S. copyright law (Title 17, U.S.C.). Use or reproduction of materials protected by copyright beyond that allowed by fair use (Title 17, U.S.C. § 107) requires permission from the copyright owners. The use or reproduction of some materials may also be restricted by terms of University of California gift or purchase agreements, privacy and publicity rights, or trademark law. Responsibility for determining rights status and permissibility of any use or reproduction rests exclusively with the researcher. To learn more or make inquiries, please see our permissions policies (https://www.lib.berkeley.edu/about/permissions-policies).