With talk of edge-to-cloud computing, industrial Internet, Internet of Things (IoT) and other buzzwords, what exactly is General Electric Co.’s Predix platform, how does it work with the industrial Internet, and how is it better than traditional cloud platforms?
Nikhil Chauhan, director of product marketing, Predix, at GE Digital, sat down with Jeff Frick (@JeffFrick), host of theCUBE, from the SiliconANGLE Media team, for a special on-the-ground segment at the GE Digital compound in San Ramon, CA, as part of the GE Digital Innovation Day, to break down Predix.
The Edge-to-Cloud Continuum
With a trillion dollars in industrial assets, including locomotives traversing diverse terrain, jet engines in the sky and water injection pumps in the middle of the desert, the need for machine log data collection is extremely important for their ongoing monitoring and maintenance. These assets are on the “edge” of the network, and when data is collected there and sent to the cloud, this is what is known as “edge-to-cloud” computing.
The gigabytes of data being collected on each asset, and the sheer number of assets in the field, comprises massive amounts of equipment data being generated and transferred from all corners of the earth. Predix is designed to not only communicate with this machinery in real time, but also to provide a platform that developers can use to perform virtually any kind of advanced analytics on the data obtained. This allows customers to manage the remote machinery much more closely, thereby drastically increasing efficiency and saving their company an enormous amount of money.
“Predix is a complete edge-to-cloud platform,” said Chauhan. “The platform has technologies which can run at the edge of the devices, in the cloud, and be able to orchestrate application logic, data and control at radius levels.”
Why Cloud Is Important to Industrial Internet
When dealing with industrial assets on the edge of a network, the equipment must be able to communicate with the network and other edge machinery in real-time. With so much data being transferred across the network, latency becomes a serious issue. To solve this, Predix places compute power on the asset itself, where some of the less complicated analytics can be done on-board, thereby reducing data transfer and thus reducing latency, according to Chauhan.
By having far-flung industrial equipment handle some of the data processing itself, as well as work hand-in-hand with the main computing and analytics being done in the cloud, the system is able to shorten latency times down to a manageable level. Thus, this very broad network is able to communicate edge-to-cloud in virtual real-time.
“In mission-critical applications, you have to have systems which have to be deterministic, they have to run in real-time or near real-time situations, and you have to have very short, small latency time periods,” said Chauhan. “Edge computing working in tandem with cloud solves and addresses these limitations.”