Small Form Factor Computing Rises to Teraflops
The IIoT platform is the glue that pulls everything together and enables new insight.
The Industrial Internet of Things (IoT) – or Industry 4.0 – benefits from increasing performance and decreasing the cost of computing. Since wirelessly sending information can get expensive, it’s essential to have powerful enough “edge computers” to cull poor data and compress relevant data before sending it to powerful cloud servers for processing. High-performance computers are necessary for IIoT where teraflops of computing power are common in a small form factor computer the size of a brick, performing at the edge of the physical process. One tiny Intel processor on a small chip can make it all happen. One FPGA can fit computing to a specific purpose.
Industries are rapidly going wireless. IIoT platforms can live in factories and extend to a cloud services provider, where immense computing power and storage resides. As Artificial Intelligence (AI) and many software applications perform analytics, other software applications are made available to the IIoT platform through Application Programming Interfaces (APIs). The flow of data can come from sensors or other sources, and data sent wirelessly to the cloud.
There is almost always a cost for data flowing across the connection between the two IIoT platform components: the physical process and cloud services. Edge computing takes place on the physical plant side and squeezes data into only what’s necessary to store and analyze on the cloud side of the IIoT platform. At multiple petabytes of data, it’s wise to pre-process data before sending it to the cloud to keep network costs down.
Once in the power of the cloud, one can perform analytics on the data or feed it into AI. IIoT can extract insight into what is happening within the physical process. Results can then be used for improving logistics, harnessed by management personnel, analyzed by data scientists, enable preventive maintenance by operators, and so forth. Alas, IIoT still requires a lot of human intervention. However, IIoT is not a huge machine where magic answers pop out of the data. Data scientists are key to teasing insight out of the data for everyone working in the physical process.
Operators might modify the physical process. Logistics teams might change the amount of raw material coming in or find new ways to ship products by analyzing data from every facet of the physical process. The IIoT platform is the glue that pulls all of it together and enables it to happen.
Lynnette Reese is Editor-in-Chief, Embedded Intel Solutions and Embedded Systems Engineering, and has been working in various roles as an electrical engineer for over two decades. She is interested in open source software and hardware, the maker movement, and in increasing the number of women working in STEM so she has a greater chance of talking about something other than football at the water cooler.