Analysing Wi-Fi in near real time

iPass is using Databricks’ Unified Analytics Platform and machine learning capabilities to monitor Wi-Fi hotspots in near real-time, and ensure mobile devices are connected to the most accessible hot spot measuring speed, availability, performance, security and location. With Databricks’ Unified Analytics Platform, iPass accelerated time-to-market of its Veri-FiTM suite of mobile device analytics and big data services.

  • Thursday, 8th March 2018 Posted 6 years ago in by Phil Alsop
iPass’ SaaS solution provides enterprises, operators and brands advanced mobile connection management, mobile connectivity and mobility intelligence.  As a part of their service, iPass offers secure, single sign-on access to the world’s largest Wi-Fi network, with more than 64 million hotspots in airports, hotels, airplanes, and public spaces in more than 160 countries and territories across the globe. Analyzing the reliability, performance and other aspects of these millions of Wi-Fi hotspots is a daunting task, and iPass was challenged to keep pace with the ever growing, unregulated Wi-Fi environment. iPass knew it needed to move from an on-premise Hadoop system to the cloud with Apache Spark to manage and scale their data and meet customers’ expectations of 24/7 Wi-Fi access.
 
Though they were on the right path, the data engineering team found it increasingly challenging to build and maintain Spark clusters without compromising the critical tasks of writing business logic, and the delivery of saleable and scalable products. Within weeks, the team contacted Databricks and was up and running with Databricks’ Unified Analytics Platform, which eliminates the need for disparate tools, reduces deployment times to minutes, and increases the productivity of data engineering teams. The team also uses Databricks for building ETL pipelines, streaming, analytics, machine learning, data lake processing and reporting.
 
“With more than 100+ million hotspots to maintain, we were experiencing scalability, maintenance and complexity challenges with our old Hadoop infrastructure that distracted our small data engineering team from reaching their goals with real-time analytics,” said Tomasz Magdanski, director of Big Data and Analytics at iPass. “Databricks was the obvious solution as it’s equipped to collect and scale massive amounts of Wi-Fi data. We have been using Databricks in development through to production, which has allowed our data engineers to focus less on building infrastructure and more on creating new data products and providing the best Wi-Fi recommendations to mobile devices - ultimately improving the user experience.”
 
iPass has seen the biggest business benefits with quick iterative development, fast time to market (from months to weeks), easy research and development and generating new revenue opportunities with Veri-Fi. iPass also leverages Databricks’ engineering team for their knowledge and expertise when it comes to building and deploying new machine learning models.