As edge computing deployments get underway, organizations need to make their edge computing sites resilient. As the saying goes, “you’re only as good as your weakest link,” so if edge computing locations are allowed to be the weakest link in a multi-location environment, the entire network surely will suffer.
You can’t have a truly resilient IT deployment without resilience at these edge computing sites. So organizations have to harden these sites with best-in-class technology as they do at centralized and regional data centers. Organizations need redundancy, security and management control designed to prevent downtime.
The gap between downtime in data centers and edge computing sites is significant. The average data center experiences 30 minutes of downtime yearly compared to 29 hours – or 1,740 minutes – at edge computing sites. Now multiply the edge computing sites by dozens, hundreds or thousands, and the cumulative amount of downtime is staggering. It can really put a damper on your operations, especially considering the types of applications running at these edge computing sites – business-specific processing and analytics applications requiring real-time responses.
Resilience at Edge Computing Sites is a Critical Requirement
Why is resilience at edge computing sites so critical? In modern hybrid IT environments, local edge computing creates a high-performance bridge for public and private clouds, co-located corporate data centers and local on-premise IT deployments.
The reason edge computing deployments are necessary has to do with the immediate nature of the applications in IoT environments. We’re talking applications that require real-time processing and analytics, such as mobile banking and healthcare device monitoring. In either case, latency in data transmissions can be seriously problematic.
If the data for these applications has to travel all the way to a cloud infrastructure somewhere, it is bound to experience latency. That’s why, rather than relying on centralized cloud environments for IoT, organizations are building edge computing sites much closer to where the data needs to be processed for prompt action.
The Shift to Decentralized IT Deployments
The emergence of edge computing is just another in a series of pendulum swings for IT deployments. Initially IT was centralized, with the mainframe at the heart of it all. Later on, computing was decentralized and replaced with a distributed client/server model that helped drive down costs and increase scalability. But those networks kept growing, often in unkempt ways that added costs, complexity and management challenges.
In the past decade, the tide turned back to a centralized model as clouds based on virtualized environments replaced physical servers, delivering better scalability and flexibility and lower costs. Now, as IoT takes hold, it’s clear centralization is limited in supporting applications for which the cloud was not designed.
The cloud was conceptualized for specific applications such as email, productivity applications and communications. Applications were binary and performance wasn’t a major consideration. Either they were available or not. In the world of IoT, performance is paramount for functions such as smart fitting rooms and augmented reality in retail locations, connected operating rooms and smart beds in hospitals, mobile banking and autonomous driving.
These applications are different from those developed for the cloud in several ways:
- Higher bandwidth requirements
- Lower latency requirements
- Stricter regulations pertaining to customer data management
Increased Need for Remote Monitoring of Edge Computing Sites
To increase resilience at edge computing sites, which typically are unstaffed and too often deployed in an ad hoc fashion, organizations need to understand three imperatives that help minimize downtime:
- Remote site management
- Physical security
- Standardized, redundant deployment
Access Edge Computing Resources
APC by Schneider Electric offers solutions that are ideal for edge computing sites because they address all three pillars. By deploying these remote management solutions, organizations will ensure their edge computing sites don’t become the weakest link in their IT environment, and instead, gain Certainty in a Connected World. To learn more about edge computing solutions, review our edge computing resource page, which includes white papers, videos and more to help you explore options for your business.
This article was originally published here.