Industry executives and experts share their predictions for 2023. Read them in this 15th annual VMblog.com series exclusive.
Life at the Edge is Getting More Powerful, Secure, and Efficient
By Hoon Choi, Fellow, Lattice Semiconductor
If one thing is clear moving into 2023, it’s that Edge computing is the future. Organizations are beginning to understand the value of processing data locally, and the Edge technology to do so is becoming more accessible than ever. It’s a particularly exciting time of rapid innovation, which has quickly changed how some businesses, especially those in manufacturing, approach IT.
Below are a few observations about what I believe we can expect in Edge computing in 2023.
More Intelligence at the Edge – With the Computing Power in Small Devices to Process It
With more Edge devices being deployed both commercially and domestically, more data is being collected and processed. As a single Edge device attempts to process increasing volumes of data, it requires more computing power.
Advancements in machine learning and deep neural networks have helped increase the computing power of small Edge devices so that they’re capable of meeting higher processing demands while also operating at a lower power and within safe thermal ranges. This has significant implications for industrial applications like automated robots or conveyor belts requiring ultra-low latency to operate reliably and safely. Companies need these devices running at top efficiency, but the cost of powering them and their cooling resources can be a significant challenge. To simplify thermal management and lower operating costs, more and more companies are looking at flexible computing solutions running on low power Field Programmable Gate Arrays (FPGAs) that can be programmed to perform nearly any type of computing function.
While its evolution seems to have limitless potential, it’s important to remember that automated robots and other Edge applications can only improve if they have the intelligence and processing power to do so. For those industries eagerly awaiting these advancements, they’ll want to keep their eyes open for further advancements in this space.
Greater Focus on Edge Device Security
Security is never a given. While people like to think cutting-edge technologies have cutting-edge security protecting them, that’s not always the case. This can even include the very technologies consumers depend on to keep them safe, for instance smart security cameras or other smart home devices. While the ramifications for a weak door lock can be easily understood, consumers often have trouble understanding how, for instance, an unsecured artificial intelligence (AI) assistant could put them at risk. As smart home devices learn about the user’s preferences, that data is sent back and forth between the device and a server, updating constantly. It is vital this data is treated with the seriousness it requires, since it pertains to consumers’ privacy and information about their homes. This also holds true for commercial applications like smart factories, 5G telecom infrastructure, and connected cities.
Convergence of AI and Security Features Will Enable More Powerful, Secure, and Efficient Computing
Edge computing has reached a unique point in its maturity where AI and security features of connected devices can be combined further to advance capabilities at the Edge. This will not only drive innovation for brand new Edge applications, but also make existing tasks less processing intensive. It will be interesting to see how this unfolds in, for example, the Automotive industry. By making smart cars more secure from cyberthreats and improving the performance of processing intensive features like Advanced Driving Assistance Systems (ADAS) or autonomous driving, Edge devices are turning cars into their very own efficient computing engines.
Exciting Times at the Edge
Lattice Semiconductor is proud to be a leader helping drive innovation at the Edge with our low power FPGAs and software solutions, like our newly released power efficient FPGA platform Lattice Avant that is optimized for Edge applications like data processing and AI. It equips system and application designers with reprogrammable low power, small size, and performance optimized processing solutions ideal for the ever-evolving Edge.
Without a doubt, 2023 is shaping up to be yet another massive year for the Edge computing ecosystem. We will see more intelligence processed by smaller Edge devices powered with AI computing engines. There will be a greater focus on Edge device security, and the convergence of key features will make for a more powerful, secure, and efficient Edge computing future.
##
ABOUT THE AUTHOR
Hoon Choi is a Fellow at Lattice Semiconductor, where he leads the development of machine learning technology and solutions, and domain specific accelerators and SW libraries on low power FPGAs. Previously, he developed elliptic curve cryptography-based security blocks for FPGA products and did HDCP2.2 specification and design work for HDMI products. Prior to joining Lattice, Choi obtained his Ph.D. from KAIST in Korea then began his career in the technology industry working for Samsung Electronics, NeoPace Telecom, and Silicon Image.
