A little more than 20 years ago, IBM introduced the concept of ‘autonomic computing’, which took inspiration from the human body’s ability to regulate critical functions without conscious thought. The ultimate goal of this ambitious initiative was to build self-healing computer systems that could automatically diagnose and remediate issues without the need to summon a human operator.
However, the real issue that autonomic computing was looking to solve was that of systems complexity which, at the dawn of the new millennium, was becoming especially acute as a fresh generation of Web 2.0 start-ups bet the farm that they could make all of these disparate systems talk to one another.
Interoperability and interconnectivity have since become table stakes for the modern enterprise – but these advanced capabilities don’t come without a cost. While we’ve made significant progress in terms of advancing our collective autonomic ambitions, the specter of complexity remains ever-present.
And perhaps nowhere in the IT landscape is this more evident than at the network edge, where skilled IT personnel are always in short supply and downtime is unacceptable.
Conquering Complexity IRL
One might think that the autonomous edge remains a far-flung vision of the future. But it’s actually much closer than you might think. Today, we’re seeing customers put autonomous edge systems to work across a broad spectrum of use cases — from deploying edge units on the factory floor in order to improve the performance of their heavy equipment to retail stores that can’t afford the latency of the cloud to run their critical applications.
To understand how edge autonomy is reshaping the world today, look no further than Northern Marine, a provider of ship management and marine services, who selected Scale Computing’s Edge solution to automate and remotely manage the IT infrastructure across their fleet of vessels that operate around the globe. These ships rely on a variety of critical applications such as crew management, cargo management, accounting, security, and maintenance systems to keep their crew and cargo safe and running on a 24x7 basis.
Because these ships operate in one of the most extreme edge environments — in the middle of the ocean where internet connectivity is at best highly intermittent — automation capabilities were a key requirement, especially since there is no IT staff on board to troubleshoot or resolve issues when they arise.
As Scott Mungall, Northern Marine’s IT Technical Lead explained in this recent case study, “we need to be able to sustain equipment failure and still continue to operate, as failure of a critical system can lead to detentions or vetting findings which can cause us to lose charters, or in the worst-case scenario, cause a vessel to be impounded. The ability to automate our backups using snapshots is another area that has greatly improved our resilience, and in the future we plan to use Scale Computing’s automation capabilities to update the onboard systems and software, as making manual updates to 70 ships takes quite a bit of time.”
From Automation to Autonomic
There are a couple of different aspects to automated edge deployments, which like any novel technology, span a broad continuum of capabilities.
At the low end of the automation spectrum, there is what some would classify as baseline or human-triggered automation capabilities, which includes the ability of an edge system to perform basic automated functions such as one-click rolling updates to deploy system updates and patches or notifying administrator to provision new nodes when predefined capacity thresholds are triggered and automatically expanding the cluster resource pool.
Perhaps the most common application of edge automation that we see with our customers is utilizing automated backups of their data and snapshotting their servers and VMs to ensure resilience and accelerate recovery in the event of a disaster or disruption. Increasingly, these types of capabilities have become relatively common and most modern edge deployments will include some underlying level of automation.
On the other end of this maturity spectrum is a fully autonomic, self-healing edge infrastructure in which the system itself is able to diagnose and mitigate issues as they happen, only calling out to the administrative team when human intervention is absolutely required. In this respect, the autonomic edge of the future will be driven more by software intelligence than hardware integration – advanced capabilities such as AI-driven self-optimization will undoubtedly play a significant role as autonomic edge systems will require the ability to leverage data in real-time from the underlying systems and components in order to accurately predict and preempt outages.
Edge Autonomy Begins with Zero Touch
One of the major obstacles hindering large-scale enterprise edge deployments is the ability to cost-effectively manage multiple edge deployments as a single, cohesive unit. For instance, say you are a large manufacturer with a thousand edge systems spread across five facilities. Now let’s imagine there is a critical patch that needs to be applied to each of these servers, and applying that update takes about 12 minutes of administrative time, plus downtime for each server.
Here, a fairly ordinary type of IT event is now slated to absorb 200 hours of administrative time (ignoring any need to physically touch the servers, or the risk of introducing unintentional human error at each step), plus potential downtime of data collection and perhaps even the machinery tied to the infrastructure. A simple update like this could at best be a $50,000 or more endeavor – provided of course that it all goes smoothly.
Fast forward another few years when data collection capabilities have increased by a factor of two or three-fold per year. What happens if we aren’t talking about 1,000 devices, but 10,000 or even more?
This is why Zero Touch Provisioning, which enables a centralized IT staff to remotely deploy and manage their edge-based systems from a single console, promises to be a true game changer and a foundational component of the autonomous edge.
Zero Touch Provisioning will also enable customers to automate and streamline the process of provisioning an edge node – or even an entire cluster – meaning that an administrator at corporate headquarters can simply drop a unit in the mail to one of their edge locations and the recipient would only need to plug the unit in as the entire provisioning of the system would be configured via a cloud edge portal.
It’s important to remember that the edge is not a single technology, but a collection of technologies working together to connect applications and data to users where they are needed most. Ultimately, the promise of edge autonomy is for the edge to manage, diagnose, and heal itself. And every day, Scale Computing and our customers are moving closer to realizing this grand vision.
To learn more about how Scale Computing is driving automation at the network edge, download the free Gorilla Guide: “Enabling IT at the Edge”