icon-play icon-normal-size icon-expand icon-pause iconExit Click for search iconTarget iconCheckmark iconEngine iconTruck linkedin_icon twitter_icon arrow arrow-left arrow-right close iconWorkforce iconAudit iconEmergency icon-pin icon-dots icon-act playbutton pause-button


Episode 6: Making Sensor Data Actionable and Valuable at Scale

Share this Video

Episode 6: Making Sensor Data Valuable and Actionable at Scale
Series: Making Sense of Sensor Data

[00:00:13] We’re headed to a world of trillions of sensors. It’s already in the hundreds of millions and sensors are getting aggressively cheaper; they’re getting smarter.

[00:00:22] You start to get to look at scale in a different way than just traditionally scaling a cloud platform.

[00:00:29] If you receive a piece of data you have to determine is it relevant to any decisions are going to make. If it’s not relevant and you have to determine is it valuable for some other service.

[00:00:37] The value of data in general comes from what it’s used for. It’s about making it classifiable and actionable.

[00:00:44] The role of artificial intelligence and the role of machine learning is going to be progressively invasive, but none of that matters if you’re not able to cope with the scale and the velocity of the data. So we start there.

[00:00:57] This is where processing can come into play. It’s all very configurable so that you can handle this scale problem with all your available resources and not just sending everything constantly to the cloud.

[00:01:09] The difference with sensor data is that time series data. It’s not structured record oriented data. The meaning of the data is not only in the initial data points themselves but in the trendline associated with seeing what that device is reporting over a period of time.

[00:01:27] Many companies have made decisions on the analytics side but the data that feeds analytics is equally as important. And so that’s where Sixgill sense helps to bring in very granular actionable data that can feed the power of many of these analytics platforms.

[00:01:42] We can take contextual record oriented data that’s necessary to inform an automated logic and separate that from the time series data that is going to be direct and acted on.

[00:01:55] We have gone into companies like Amazon and Azure and Google Cloud and started to look at integrating our services directly on the edge of those clouds. We can capture all the data that’s coming from these billions of devices as it reaches that funnel the service that we’re providing not only aggregates data but we do the cleansing and normalization of it.

[00:02:17] We’re going to make the system progressively more intelligent. We’re going to make a progressively more secure. We’re going to do all things that you have to do on the backend to make it something that developers can forget about just simply adopt and rely on.

[00:02:29] We tie in economics to sensor data to moving the decision services from the edge to the cloud based on particular application requirement. I could create an opportunity to make all sorts of decisions after the economics of things that today don’t have an economic value proposition.

[00:02:46] We’ve built out the sensor data agnostic platform from the start. We take in all of that data as we connect these sensors in a scalable maintainable way. It’s going to handle all of the data problems that you’ll have if you aren’t thinking about that ahead of time.