In this DemoFriday™, CALIENT Technologies, InMon, and Cumulus Networks demonstrate a programmable hybrid data center network that detects and offloads elephant flows from a Layer 2/3 network fabric to a pure photonic Layer 1 optical circuit switched network.
CALIENT, Cumulus and InMon SDN Optimization Demo Synopsis:
Modern cloud-scale data centers support a variety of traffic patterns, from traditional, short-lived mouse flows to more persistent elephant flows. While mouse flows are typically associated with “bursty,” latency-sensitive applications, elephant flows include traffic from virtual machine migrations, backup and data migrations, and Big Data applications such as Hadoop. The majority of flows within a data center typically are mouse flows, but the majority of data belong to a few elephant flows.
Unfortunately, elephant flows tend to fill network buffers end-to-end, which adds significant queuing delay to the mouse flows that share these buffers. The more latency-sensitive mouse flows suffer from increased latency caused by elephant flows.
In this DemoFriday™ from September 19, CALIENT, InMon, and Cumulus Networks demonstrate a programmable hybrid data center network based on real-time traffic patterns and specific application priorities. Offloading the elephants facilitates more efficient and cost-effective delivery of the flows while freeing the Layer 2/3 network to focus on handling the short-lived mouse flows it services very effectively.
The demonstration showcases a controller developed by CALIENT that uses InMon sFlow and sFlow RT components to detect elephant flows between bare metal top-of-rack switches running Cumulus Linux. The controller offloads and reroutes the flows through a CALIENT S320 optical circuit switch.
Watch the SDN Optimization DemoFriday™ full presentation featuring CALIENT, InMon, and Cumulus Networks or take a look at these additional resources:
Disclaimer: Statements and opinions expressed in webinars and other related materials are those of the authors, presenter, and organizations they represent; and do not necessarily reflect the views or policies of SDxCentral, LLC or its publishers, editors, employees, members, partners, or sponsors.
SDxCentral, LLC makes no representation concerning and does not guarantee the source, originality, accuracy, completeness or reliability of any statement, information, data, finding, interpretation, advice, opinion, or view presented. SDxCentral will not be held responsible for any claim, loss, damage or inconvenience caused as a result of any information within this site, or any information accessed through this site.
Notice: Pursuant to the SDxCentral Terms of Service, by viewing or downloading sponsored content such as pdfs, podcasts or videos, or signing up for our webinars, you agree to allow SDxCentral to share your contact information with the sponsor.