Edge devices for the IoT are typically small in size, are run by a low-end microcontroller, include small amounts of RAM, Flash and ROM, and have low power consumption. Off-the-shelf real-time operating systems (RTOSs) are not geared toward this environment, and hence are unable to solve many of the problems faced by the IoT designer.
In this talk, Steve Jordan will propose a different approach for an IoT RTOS. IoT edge devices first and foremost require an Internet-centric connectivity OS that can handle the Internet's non-deterministic nature. For instance:
Thus, (1) an IoT OS must exhibit the characteristics of a state machine architecture since the Internet is more connectivity-driven than what is supported by conventional RTOSs, and (2) an IoT OS must be architected for the lowest node power possible. Steve Jordan will elaborate on how these requirements can be achieved.
Steve has consulted to technology businesses for over 20 years, with a focus on developing successful growth strategies for both Fortune 100 companies and venture funded startups. He was a Partner with R. B. Webber and Co. (RBW), a leading Silicon Valley technology strategy management consulting firm, as well as an Engagement Manager with McKinsey and Co. in its technology practice. He earned a BS in EE and a BS in Economics at MIT, and an MBA at Harvard Business School.
Big Data SIG Pre-Meeting (6-7 pm)
Chairs: Shashidhar Sathyanarayana and Brian Berg
Location: Aristotle Room
This talk will start with a brief overview of the Big Data ecosystem, and will include an actual use case of credit card fraud detection with sequence mining on a Big Data platform.
How Hadoop gets used to build a machine-learning model using historical transaction data will be shown, as well as how Storm processes real time transaction data and identifies potentially fraudulent transaction sequences making use of the model built by Hadoop.
Pranab Ghosh is a Big Data Consultant who has worked with a myriad of platforms and technologies including mainframes, real time systems, Java, enterprise applications, big data and cloud technologies. His focus for the last several years has been Hadoop, Hive, Storm, NOSQL databases and the surrounding big data ecosystem.
Pranab maintains a blog on topics such as Big Data in general, as well as machine learning solutions in Big Data. He is a graduate of IIT, MIT and the Univ. of California.