Broadview brings visibility to networking
November 2015 – The shift to more cloud based processing and traffic management through softwar defined networking (SDN) and network function virtualization (NFV) are encountering increasing challenges in providing optimization for the data traffic. These challenges include having to cross hardware from multiple vendors – Cisco, HP and Juniper – that may reside in the same location and network. Broadcom has recently introduced a new software product to help with traffic and data optimization activity by providing visibility to the data flow.
The Broadview software is optimized for the Tomahawk chip in the StrataXGS switch family. This 100Gbps switch is being quickly adopted for top of rack applications by a variety of vendors. Due to the diversity of applications for these high speed devices and the variety of hardware implementations, the Broadview software has a telematics API that covers key features across vendors. The result is the extraction of data across the entire network stack including the top of rack, data plane and core.
Boradcom Broadview network analytic software
The software can perform rich congestion analytics as well as packet & flow visibility like an advanced wireshark. Issues that can benefit from the interpretation of results of the program include analysis of “Bursty” traffic patterns that are prevalent in cloud data centers that have high levels of peak utilization. The workloads they highly varied and non-uniform in nature. Examples of these diverse workloads include use of MapReduce and distributed file systems in Big Data analytics, distributed caching related to high performance transaction processing, streaming media services, and other demanding, high bandwidth computing processes.
The software is platform and protocol agnostic and supports many of the open standards for networking. The software provides a faster TTM with reference implementations for OpenDaylight, OpenStack, Ganglia etc. Many of the challenges for 25G-100G networks only occur at normal operating speed, so the software can be programmed to report based on a trigger or when the system is running at speed, even in a top of rack application.
