Edge Computing’s Need for NVMe Computational Storage, Part 2

Apr 25 2019

Scott Shadley

Our last blog discussed why computational storage was a good fit for edge computing from a power, cooling, and space perspective. If putting the right amount of compute power and storage into a small space was the only challenge that edge computing faced, any number of single-board computer (SBC) manufacturers could have already solved this problem. These are companies that adapt existing technology to ruggedized environmental demands – think military, aerospace, and industrial applications as an example of the types of punishing environments that these companies build SBCs and other products for. So what is the real issues that makes building edge computing so difficult today?

The simple answer is the amount of data that the Internet of Things (IoT) generates today is fundamentally different than the amount generated by embedded use cases even a decade ago. According to Cisco, IoT devices generate 5 quintillion bytes of data every day (that is 5 million terabytes of data), and this will continue to grow as the number of IoT devices grows to 30 billion connected devices by next year (2020), and 80 billion devices by 2025. Then consider that in all likelihood 10% of these devices generate 90% of the data – we are talking about datacenter-scale traffic levels. Think about today’s modern airliners, which can generate up to a terabyte of data per flight. Even taking small snapshots of this data, which can reduce it to below a gigabyte per flight, is still far too much to transmit in-flight, so this massive amount of data must be analyzed at the edge if there is any chance of utilizing it in real time.

This is why edge computing/IoT problems are often like enterprise “big data” problems, just in a much smaller envelope. Yesterday’s embedded computing and storage technologies simply aren’t made to handle this amount of data. Computational storage on the other hand is built EXACTLY for this type of problem – the real-time analysis of petabyte-scale data sets. This is yet another area that the TECHunplugged article discusses in further depth. Let me know if NGD Systems can help you with your edge computing/IoT problem. Thanks!

NGD Processor chip.

355 Goddard, Suite 200
Irvine, CA 92618

For Sales Inquiries:


For General Inquiries: