Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Aeris + Cassandra: An IOT Solution Helping Automakers Make the Connected Car a Reality


Published on

Drew Johnson, Vice President of Engineering, Aeris Communications

Published in: Technology
  • Be the first to comment

Aeris + Cassandra: An IOT Solution Helping Automakers Make the Connected Car a Reality

  1. 1. Thank you for joining. We will begin shortly. Aeris + Cassandra: An IOT Solution Helping Automakers Make the Connected Car a Reality Drew Johnson, Vice President of Engineering, Aeris Communications
  2. 2. All attendees placed on mute Input questions at any time using the online interface Webinar Housekeeping
  3. 3. Who Am I? 4Confidential Drew Johnson Vice President of Engineering • More than 20 years of experience at both large and small companies • Started career at Compaq Computer working on the first generations of mobile computing devices • Prior to Aeris, Director of Engineering at Austin mobile start-up SoloMio and VP of Engineering at mobile and messaging software company Openwave • Holds a Masters degree in Artificial Intelligence from the University of Illinois and has more than 20 patents
  4. 4. Agenda 5Confidential • About Aeris Communications and IOT market trends • The anatomy of an IOT PaaS • Customer challenges in delivering IOT/M2M solutions • The Aeris AerCloud PaaS offering • Evaluating database technologies • Case Study: How automotive OEMs leverage Aeris + Cassandra • Moving forward with Cassandra • Q&A
  5. 5. The Future is Here – IoT Trends Confidential 6 Volume of IOT data by 2018 90 percent 400 zettabytes 30 billion IOT data in the cloud by 2020 Connected IOT devices by 2020
  6. 6. About Aeris Communications Enterprise/MNO Connectivity-as-a- Service Cloud-based IoT Applications and Analytics PaaS Automotive Applications/ Services Aeris IoT Platform
  7. 7. The General IOT PaaS Anatomy Confidential 8 Requirements • Collect and store large amounts of data (time-series, location, sensor data) • Reliable / secure • Minimize data transfer • Process in NRT for alerts • Handle batch/deep analytics • Share data in a secure manner • Rapid prototype to production scale • Cost effective Acceleration Location Speed Business Value Traction
  8. 8. Customer Challenges w/o PaaS Confidential 99Confidential • Customer developers focused on infrastructure rather than differentiating features - suffering frequent downtimes • Data intake volume (Aeris: Grew from 1B transactions/year to almost 1B per day!) • HIPAA / Data location compliance • Need to derive more value from the data being captured • Wild Wild West on device protocols – have to adapt easily
  9. 9. How Aeris AerCloud Works 10Confidential
  10. 10. Confidential Aeris Technology Map Web Services MQTT Data Storage Indexing Analytics Location Geofence and Search Queuing Netty Tomcat Netty Aeris Cassandra Aeris Aeris RabbitMQ HTTP Push Aeris
  11. 11. Key Database Requirements 12Confidential Strong real-time search and query capability Support high reliability and security Structured & unstructured real-time data Flexible data model with affordable scalability
  12. 12. Why Not Relational Databases Confidential 13 • Costly/Complex to scale – especially w/ HA • Manual sharding at scale is a pain • Inflexible data model – can’t handle IoT variety of data
  13. 13. Why NoSQL and Apache Cassandra 14Confidential • Purpose built for IOT & time-series data • Optimized for storage & retrieval of time-series data • Integrated analytics with Hadoop connector • Fantastic write performance (2B MQTT/day @ 8 nodes) • Simple to manage (No master/slave, shared storage, manual sharding) • Supports 100% uptime through masterless architecture & multi-datacenter replication • Flexible schema allows for faster time-to-market • Commodity storage/instance storage for linear & predictable scale • No lock-in to a specific cloud provider
  14. 14. Confidential 15 Aeris has done especially well in Automotive Aeris Automotive
  15. 15. Confidential 16 • Aeris handling 1M+ vehicles for an OEM • Recently identified key anomaly due to software change • Same analytics engine could identify other vehicle system anomalies OEM Case Study
  16. 16. Moving Forward Confidential 17 Next up for Aeris – key customer areas: • Vehicle telematics – expanding use cases for both make-money and save-money • New verticals – avionics and fleet Our 2 cents: Plan for Big Data and Data Portability from the start! (Cassandra/AWS > AWS alone)
  17. 17. Q/A Input questions at any time using the online interface
  18. 18. Thank You! 19Confidential