Edge computing can become ‘Smart Computing’ with automation

Categories Edge Computing News  |  Guest Posts
Edge computing can become ‘Smart Computing’ with automation

This is a guest post by Ashwin Mistry, co-founder and COO at Taubyte.

From bare-metal servers to containers and today’s serverless approach, software stack abstraction has significantly reduced operational cost, time to market, and product lifecycle. In turn, software lifecycles have reduced, prompting a DevOps culture to automate the manual task of handling the system and networking layers needed to run the applications.

With the emergence of edge computing fueled by IoT and 5G, the software operational overhead is growing exponentially and the current code-based automation, more commonly referred to as DevOps, is not scaling adequately to meet demand. That is, as IoT devices on-board onto the internet in record numbers, so does the variety of software applications. This translates to a great deal of custom code that requires upkeep based on industry trends and economic activity.

According to a white paper by the DevOps Research and Assessment (DORA) research and consultancy, a product line valued at $100M can cost about $5.6M to automate initially. This includes the cost for on-boarding, tools, and engineering salaries to transform the existing product line to a DevOps culture. Much of this figure, $2.35M, will be an annual recurring expense with the lion’s share going to automation tools. Note that these expenses are for a cloud-based solution, where data centers are regional, and far fewer to deal with than in an edge-based solution as conveyed in the following.

Legacy cloud providers operate in multiple regions and sometimes offer hybrid solutions with other cloud providers, which ultimately pose a challenge in scaling. Coordinating DevOps in this scenario, where regional data centers are limited in number (within 20 to 30 as of 2020), is already proving to be a struggle. As edge computing services begin to build traction and developers are deploying in a scenario where data centers are significant in number (in the 1000s), there needs to be an alternate solution that is smart and easy to integrate with.

For IoT and IIoT product manufacturers and solution providers, an intelligent method is needed to reduce the software “operational” overhead, so they can home in on their core competencies. Consider the following from Mark Thiele, founder, and CEO at Edgevana:

“As ‘smart’ everything becomes more prevalent and tools like VR are integrated into local shops and businesses to save on training and to improve customer engagement, the volume of data created and used will skyrocket. Gartner estimates that 75% of all data created in 2025 will be created outside of the on-premises or cloud data center. We’ll be creating so much data that centralizing it will be cost-prohibitive. Building solutions that solve for localized analytics at the edge, will, in the end, reduce costs and, by accident, also deliver applications with low latency, which provide more engaging customer experiences.”

To address this challenge and take advantage of the related benefits, a Smart Computing approach is needed to reduce or eliminate the associated overhead in software development; i.e., a NoOps platform that enables scaling software to the global infrastructure catalyzed by edge technologies.

What does this mean for businesses? For one thing, it signifies that they can ship software to their customers expeditiously, enabling them to innovate and iterate rapidly, thereby fostering a culture of outperforming the competition and winning in the marketplace. Ultimately, this frees up company resources and allows them to focus on their business logic.

“Smart Computing is a distributed computing platform which streamlines running business logic on a global computing infrastructure that extends from data centers to connected devices,” according to Samy Fodil, Founder and CEO at Taubyte.

Unlike the DevOps computing model, where scaling is an issue, and, as mentioned above, labors to keep pace with the record number of emerging data centers, a NoOps approach is inherently scalable to any number of data centers. This is Smart Computing powered by a truly NoOps platform!

About the Author

Ashwin Mistry is co-founder and COO at Taubyte. Taubyte is a startup offering a distributed edge computing platform that eliminates software development, deployment and routing complexity at the Edge.

DISCLAIMER: Guest posts are submitted content. The views expressed in this blog are that of the author, and don’t necessarily reflect the views of Edge Industry Review (EdgeIR.com).

Article Topics

 |   |   |   |   | 

Comments

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Edge News, Analysis & Opinion

Featured Company

Edge Market Reports

Explaining Edge Computing

Edge Computing Events

EdgeIR.com

Latest News