At MEF18 in Los Angeles on October 29, 2018, Accedian launched SkyLIGHT DataHUB IQ service analytics. DataHUB IQ combines and correlates highly granular active and passive monitoring metrics from the SkyLIGHT portfolio. Machine learning capabilities are used to detect anomalies in baseline performance, providing an insight into network behavior. At MEF18, Accedian won the award for best service analytics technology solution.
Next-gen big data analytics are required for 4G and 5G
With 4G today and 5G on the way, service assurance is due a refresh. Historically, service assurance sampling has been conducted in intervals of minutes to hours. Packet loss ratio over the more lengthy intervals can be within an acceptable range over the time frame of the whole interval. The weakness of this approach is that very short-term bursts in traffic can cause short-term packet loss. A small number of packets lost with the ensuing network congestion can cause a VoLTE call to be dropped. A negative user experience event can occur, immediately detected by the user but potentially undetected by the network operator. Accedian's SkyLIGHT tool was designed to measure network performance with fine-grained precision to identify the issues caused by such short-term bursts in traffic.
SkyLIGHT will generate vast amounts of data. In the advanced telemetry era, legacy data tools and procedures will no longer be sufficient for processing the big data load. The processing challenge is to find the underperforming network needle in the big data haystack. DataHUB IQ is designed to quickly ingest massive quantities of event data and support low-latency queries of the data. The new big data analytics approach will need to detect network performance anomalies; machine learning will be needed to identify network deficiency and predict future deficiency scenarios. Advanced telemetry may produce a vast number of alarm conditions. Advanced correlation techniques will be required to distill this down to root cause conditions that can be displayed on an insightful dashboard.
The Colt-Accedian Machine Learning-Based Elastic Networks proof of concept (PoC) provided a glimpse at MEF18 into a future defined by closed loop automation. SkyLIGHT supplied telemetry data to Colt's SDN orchestrator. To alleviate the PoC's test network congestion issue, Colt's orchestrator increased the bandwidth without the need for human intervention.
Accedian has identified the next-generation multivendor, multidomain service assurance need and responded with a big data analytics approach to process a sea of telemetry data. It is successfully pivoting from NID-TWAMP expert to strategic network analytics partner.
Straight Talk is a weekly briefing from the desk of the Chief Research Officer. To receive the newsletter by email, please contact us.