Post-Hadoop Network Traffic Analytics
Post-Hadoop Network Traffic Analytics
Big data platform offers billions of available analyses on trillions of data points in seconds.
Join the DZone community and get the full member experience.Join For Free
Kentik, the network traffic intelligence company, today announced the release of significant enhancements to Kentik Detect, its big data network analytics solution. New features include multi-dimensional traffic analytics, enabling access to billions of possible analyses operating on trillions of instantly accessible data records, new traffic flow visualizations, and support for network performance metrics.
“Kentik is the first network traffic analytics solution built from the ground up on a big data platform, offered both as a multi-tenant SaaS as well as an on-premises deployment,” said Avi Freedman, Kentik CEO. “Our latest enhancements illustrate the power of the Kentik platform to deliver analytics capabilities previously unavailable to anyone but webscale giants. In 15 minutes, customers can get from registration to production use of big data analysis that goes beyond Map/Reduce and streaming analytics-based approaches, with no appliances or software deployment required.”
The Kentik portal updates, available immediately, offer multi-dimensional analytics, which provides users with instant access to any one of billions of possible network traffic and performance analyses. Users can select, group, order, filter and visualize several data dimensions from dozens of data fields in mere seconds, even when operating on trillions of data records.
Enhanced visualizations include time-series line, bar and stacked line charts, comparison bar charts, and traffic flow charts. Users can analyze traffic based on expanded metrics that address traffic volume, such as bits, packets and flows per second; endpoints, such as unique source and destination IP addresses; and network performance, such as TCP retransmits, jitter, and round-trip time.
Kentik's instant multi-dimensional analytics contrasts with traditional big data analytics platforms such as those based on Hadoop’s Map/Reduce. Hadoop, the best known big data technology, requires a lengthy process to prepare data for analysis, and full scans of data to provide analytics. These approaches can take hours to complete, making it too slow for operational use cases. Alternate approaches involve creating pre-defined aggregates or “data cubes” comprised of multiple data dimensions, with fixed dimensions and limited granularity.
Kentik ingests network data at Internet scale--hundreds of billions of records per day--in real-time, and allows users to run multi-dimensional queries and receive visualizations in a few seconds. This allows Kentik to support key network visibility use cases at greater scale and detail than previously possible.
"As the world’s first endpoint-aware application delivery solution, we live by our ability to assure the best user experience for web and mobile applications,” said Justin Fitzhugh, Vice President of Technical Operations at Instart Logic. “Kentik gives us an edge by providing detailed real-time insight into the huge volume of application traffic that flows across our global infrastructure. The visibility we gain from Kentik is a key component in delivering world-class performance to businesses that depend on cloud application performance."
Kentik makes network operations more intelligent, providing intuitive visualizations for network planning and BGP peering analytics, thus improving user experience and network cost efficiency. Kentik’s analytical muscle makes it the ideal DDoS detection platform since it can scan massive volumes of data on a network-wide basis and alert on both attacks and operational anomalies in seconds rather than the multi-minute timeframes of traditional traffic analysis solutions.
"Network data is big data," said Jim Frey, Kentik VP Product. "Without scale, granularity, flexibility and speed of analysis, it's impossible to get actionable intelligence from it. Kentik unlocks the value of network data for network managers and operators, so they can improve service delivery and focus on improvements and innovation rather than reactive firefighting."
Opinions expressed by DZone contributors are their own.