Cisco Employs Sensors to Bring Machine Learning Analytics to the Data Center

Mike Vizard
Slide Show

The 5 Best Methods for Drawing Insight out of Machine Data

Cisco has announced it is combing machine-based learning algorithms, analytics and sensors to provide insights into data center operations in real time.

Ishmael Limkakeng, vice president of marketing for Cisco Systems, says the product they’ve dubbed Cisco Tetration Analytics provides an unprecedented amount of visibility into data centers by pulling data from both host and network sensors. It can then both correlate that data against policies defined by the IT organization and offer suggestions for how to improve overall performance and regulatory compliance.

Limkakeng says Cisco Tetration Analytics makes it possible to record everything that has occurred inside the data center. In effect, Limkakeng says, Cisco Tetration Analytics is a “time machine” for the data center.

Delivered as an appliance based on Intel Xeon-class processors, Cisco Tetration Analytics is one of the first tangible Cisco efforts to turn the network into a sensor. As part of that effort, Limkakeng says, Cisco Tetration Analytics is pulling data from layers four through seven of the network. The first iteration of Cisco Tetration Analytics will have software sensors to support Linux and Windows server hosts, while hardware sensors are embedded in the ASICs of Cisco Nexus 9200-X and Nexus 9300-EX network switches. The analytics software itself is running on an instance of Apache Spark running on top of Hadoop that Cisco has customized to support real-time analytics.

Information gathered by those sensors can not only be fed into the Cisco Tetration Analytics platform, it can also be used to inform applications that tap into the Cisco Application Centric Infrastructure (ACI) architecture for creating software-defined networks and data centers.

Advances in machine learning algorithms and real-time analytics that can be cost effectively delivered via an appliance give IT organizations real-time visibility into application flows across the data center. The issue now is figuring out what to do with that data to actually enhance both application performance and overall data center reliability.



Add Comment      Leave a comment on this blog post

Post a comment

 

 

 

 


(Maximum characters: 1200). You have 1200 characters left.

 

 

Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.