Internet of Things (IoT) Databases and Analytics Powered by In-Memory Computing
Gartner expects the Internet of Things (IoT) to have over 20 billion connected things by 2020. This many connected devices transmitting information will require an enormous amount of processing to derive value from this data. To cope with this rapid growth of the Internet of Things, successful IoT database platforms will need a data architecture which leverages in-memory computing. These architectures will address the significant challenges in terms of speed, scalability, variable workloads, and other issues created by IoT applications.
The Architecture of HTAP for IoT
The architecture of the Hybrid Transactional/Analytical Processing (HTAP) component of IoT analytics and transactions is similar to the Lambda architecture defined by Nathan Marz for Big Data applications. It takes advantage of both stream- and batch-processing methods. The Lambda architecture includes the following layers:
- A high-speed layer — a real-time processing and transactional engine (typically something like a caching system and a compute grid, such as Redis and Spark)
- A batch/storage layer— data storage with an analytical or historical processing engine, such as Hadoop with Hive
The Lambda architecture of the HTAP component of the Internet of Things deals with event-stream processing, fast analytics, and storing data for advanced and long-term historical analysis, when necessary.
For companies just getting started with IoT, combining multiple technologies can require a significant investment in terms of skill set. While it is possible to find people who know each of a variety of technologies, it is not easy to find people who know all of them. There is a lot of complexity involved.
The GridGain® in-memory computing platform provides a way to simplify the HTAP architecture for IoT databases and analytics. It addresses the needs of both transactional and analytical processing and also provides persistency and event processing — all in a high-speed, linearly scalable platform. GridGain can also serve as your primary IoT database. And GridGain is just one core technology with one skill set to learn.
Internet of Things Use Cases
According to 451 Research, 65% of companies are using IoT. 69% of organizations gather data from end points and 94% of those companies use it for business purposes. The highest usage is among Utilities (92%) and Manufacturing (77%).
The IoT data comes from:
Datacenter IT Equipment (51%)
Cameras and Surveillance Equipment (34%)
Smartphones and End Users (30%)
Buildings and Other Structures (21%)
Environmental Sensors (15%>
Factory Equipment (14%)
Automobiles/Fleet Equipment (11%)
Retail Operations (8%)
Medical Devices (7%)
No results found
The Internet of Things (IoT) is more than a bunch of sensors. Sensors and embedded devices gather data about the surrounding environment, but what you do with this data is what truly matters. A bunch of sensors won’t optimize your business. Collecting and analyzing the data that those sensors produce will.
Internet of Things (IoT) initiatives in financial services require greater performance and scalability than existing database architectures allow. This webinar will explain how to use the GridGain in-memory computing platform to overcome the performance and scalability limitations of existing data infrastructures in order to build successful IoT solutions in financial services.
Over 30 billion devices will be wirelessly connected to the Internet by 2020.
Internet of Things (IoT) applications have three key components: the devices, the networks connecting them, and the analytics that use the generated data. Before data is valuable, it must be converted to actionable information through complex processing and correlation algorithms. As the number of devices increases, so does the volume of data, giving rise to scaling challenges. In-memory data fabrics such as Apache® Ignite™ address the scaling and real-time processing requirements of IoT use cases.