Bringing Massively Parallel Processing to IoT Edge Analytics

Cisco ParStream reports sub-second responses to petabytes of data.

Using geo-distributed edge analytics and massively parallel processing, Cisco ParStream claims query response times up to 30 times as fast as its peers. (Image courtesy of Cisco).

Cisco ParStream is an Internet of Things (IoT) analytics platform that utilizes edge analytics and massively parallel processing (MPP). The platform boasts that its MPP architecture allows for sub-second response times, even with petabytes of IoT data.

Distributing Data for Faster Analysis

One of the ways in which Cisco ParStream gains an efficiency advantage is through the use of geo-distributed analytics servers. This setup processes IoT data in up to 1,000 distributed locations near the data source. Edge analytics result in a smaller footprint than centralized processing and reduce costs of ownership.

More importantly, the use of edge processing allows for real-time data processing with minimal latency. This, along with the MPP architecture, allows Cisco ParStream to handle upwards of 10 billion rows of data and stream one million rows per second. This allows users to efficiently handle massive amounts of data to obtain real-time insights.

Meeting Big Data Demands of the IoT

Cisco ParStream was recently awarded the 2016 Global Frost & Sullivan Award for Customer Value Leadership, for helping companies manage the huge quantities of data generated by IoT devices. Without an effective way to analyze and filter this data, valuable business insights could be lost.

“The combination of Cisco’s strong infrastructure and investment and Cisco ParStream’s superior IoT analytics technology has helped the latter plug the gaps in the IoT industry,” said Prem Shanmugam, a research analyst with Frost and Sullivan. “It has been especially successful in supporting high-speed data streaming, rapid query processing and analysis, and large processing capabilities.”

With IBM reporting that almost 90 percent of IoT data goes unused, Cisco ParStream’s claims of query responses up to 30 times faster than its peers is a big step toward managing big data. 

Does your IoT platform perform analytics? It should. Learn about this and other questions to ask your IoT platform provider.

Written by

Michael Alba

Michael is a senior editor at He covers computer hardware, design software, electronics, and more. Michael holds a degree in Engineering Physics from the University of Alberta.