A Federated Edge Analytics Platform for Industrial IoT
Sensors are everywhere these days from trucks, to lawnmowers to fridges and even on our bodies! Perhaps no other field has undergone as much of a transformation on account of IoT as the field of industrial analytics. Industries have mastered the use of IoT sensors across industrial assets that can be scattered across the world. The massive amount of data generated by these sensors can be effectively leveraged to optimize the performance of the assets and reduce downtime. To provide a glimpse of exactly how big of a dataset we are talking about, a GE gas turbine can generate on an average 40 million data points from over 1000 IoT enabled sensors. Now imagine you have a fleet of IoT enabled industrial assets like turbines scattered across the world. The resulting data generated is humungous! As a result two important questions pop out almost immediately. How do I store and manage this data? Moreover, how do I analyze this sensor data effectively to derive a healthy ROI?
Cloud Based Industrial Analytics
Typically, industries have started looking at the cloud as a strategy to derive value. This basically means that the wealth of data must be continuously streamed to a central aggregator for storage and analytics. This model has worked really well so far but some important issues have been cropping up lately. First, asset owners would be required to migrate sensitive data to the cloud, so data privacy becomes can be an issue. Secondly, although storing data is inexpensive, the costs of running algorithms to process that data and share insights with low latency can escalate very quickly. When you do the math the compute cost quickly adds up!
Edge Analytics
Edge computing has been proposed as a means to do local analytics near the source of data in order to provide immediate inference and insights. A recent Gartner report estimated that by 2022, more than 50% of enterprise-generated data will be created and processed outside the data center or cloud. By 2021, 80% of asset-intensive industry CIOs will be responsible for operational technology (OT) data, and will embrace IT/OT integration as a strategic imperative. Its no wonder that more and more large industrial entities are adopting edge based strategies for advancing their digital transformation efforts.
Issues with the Edge
While Edge computing works great for local data processing, it is often limited by the fact that data is siloed at individual Edge sites. As a result, in most cases insights delivered at the Edge are restricted only to the data observed at that particular Edge site! In order to obtain fleet wide insights, data must still be migrated to the cloud periodically. In fact, most of the industry experts we spoke with complained about the limitations of the edge with respect to its reliance on the cloud in order to obtain fleet wide insights.
This presents a critical bottleneck of the Edge computing paradigm. What if I have multiple industrial IoT enabled assets spread across a wide geographical area? How can I ensure that say, failure modes observed on a subset of equipment can be shared with the rest of my fleet? Wouldn’t it be amazing, if the Edge sites could talk and collaborate with each other? What if there was a way for edge sites to collaborate with each other without moving any data?
Introducing Blockalytics!
These were the exact questions that inspired us to create Blockalytics, a Federated Edge Analytics Platform for Industrial IoT. Blockalytics leverages a decentralized, federated edge analytics platform to deliver cloud quality insights. The Blockalytics platform can span multiple Edge sites and help unlock the hidden potential of data tucked away in information silos. It allows Edge sites to collaborate and build analytical models across a fleet of assets without moving any data. Blockalytics drastically reduces cloud interactions, allowing you, the user to leverage your Edge data to build global fleet wide insights! Furthermore, with our tailor-made SDK, data scientists can write and port their own analytics and machine learning applications.
The major advantage of a platform like Blockalytics is that we can ensure both the mandates of the Edge: eliminate the need for data movement and reduce compute costs. Using Blockalytics, its now possible to realize the true potential of the Edge! You get to keep your data and have your insights too! Isn’t that amazing!
Heres a short video about Blockalytics we put together to help understand how Blockalytics can add value to your enterprise.
