GPU-Accelerated Analytics for Telecommunications
Streaming Analytics Database, Ideal for Telecommunications
Ideal for Streaming Data
Transforming Telecommunications with Truly Real-Time Analytics
This demo shows the Kinetica engine powering accelerated analytics on a 14 billion record dataset of raw mobile signal events.
Network and Infrastructure Optimization
Track and visualize the real-time usage and status of telecommunications networks/towers in order to gauge performance levels and identify any bandwidth or maintenance issues. Kinetica’s flexibility to do rapidly perform predictive analytics enables telcos to identify equipment before it fails for lower service disruptions and lower maintenance costs. Cell phones can be “polled” on demand to determine the health of the cell phone network. Kinetica’s GPU cores can crunch this kind of data far more efficiently and quickly than CPUs, so results can come back in milliseconds instead of hours.
Monitor Usage and Capacity
Telcos can also use Kinetica to collect and visualize real-time data in order to identify the periods of heaviest network usage, forecast network capacity, and plan for potential network outages or short-term surges. Although a traditional RDBMS can perform these types of tasks, Kinetica’s difference is that it can process data in near real time. Kinetica is particularly well suited to location-based analytics—visualize billions of data points and display changes in real time as the underlying data changes at many orders of magnitude faster than a traditional RDBMS. By integrating a multitude of complex data sets and quickly visualizing this data, Kinetica can provide telcos with a real-time, dynamic view of their operations by providing:
- Interactive query on large and complex data sets so you can interactively explore problems.
- Fast query on data immediately as it arrives, so you can spot call drops, redials, etc.
- Location-based analytics which allow towers and cell locations to be visualized and interactively queryable on a map.
Network Usage and Call Detail Records Analysis
Kinetica’s vectorized columnar data store is ideal for large volumes of streaming high-cardinality data, such as call-data records. GPU acceleration ensures low latency between ingest and time to query, allowing for sophisticated analytics on live data. Use Kinetica to quickly identify network problems, better understand usage patterns, and more.
Telco Solutions for Smart Cities, Cars, and Homes
There is a staggering amount of IoT data that “smart cities” will need to process, and making sense of that data is at the heart of any successful smart city. As telecommunications companies support an ever increasing network of communicating devices, Kinetica is ideal for capturing that data and enabling fast insights to be made available as services to cities.
Kinetica can be used to visualize this “smart” data in real time so that cities and government agencies can gain timely insights and act on that data. Sensors can be used in public transportation systems to track traffic and real-time arrival updates. Photos of license plates can be taken and tracked to better understand general traffic patterns. Intelligent road signs could be used to communicate traffic issues with cars, and divert traffic to less congested roadways. Sensors on home water systems can be used to detect water quality and leakages. Sensors in trash containers can be monitored to plan collection times based on when the container is full. This ability results in significant savings in fuel costs, improved mileage, and reduced wear and tear on trash collection vehicles. “Smart” bridges can deploy small sensors that monitor and transmit data on a bridge’s structural integrity. Telcos can gather and analyze data the IoT data from smart cars and homes that they can then sell back to cities.
Customer Churn Reduction
There are two main limitations that can have a negative impact on telecom revenue: 1) Lack of timely insights, and 2) Delayed response time to customers. Use Kinetica to combine social media feeds, call detail records, network performance data, and call center interactions to identify declining usage and pinpoint the customers who are the most likely to defect, so that they can then address customer satisfaction issues and take appropriate steps to prevent customer churn.
Fraud Detection and Prevention
Telcos can use Kinetica to collect and analyze real-time data in order to detect anomalous behavior. Baseline normal behavior can be modeled through the analysis of customer account data, location-specific data and usage data. Kinetica can then enable the development of predictive models that can flag and proactively prevent fraudulent activities.
Subscriber Data Monetization
An emerging area that telcos are exploring relates to collecting, packaging, and selling anonymized subscriber data to other vertical markets such as advertising, retail, financial services, public services, and healthcare. Cell phone device location data can be combined with other subscriber data to provide information on what stores a subscriber visits, at what time of day, and where the subscriber lives. For example, Kinetica can be used to provide real-time heatmaps data that show who is visiting a retailer’s store and where they live. Data from a shopper’s visit to a particular store can be combined with web browsing data, which can help retailers provide customers with targeted ads, or help them determine where to open new retail locations.
More Telecoms Use Cases
Chat with one of our industry specialists for more in-depth use-cases and examples of how Kinetica is being used in telecommunications today.
We can also walk you through a demo, and set you up with a trial environment for you to experience it for yourself.