As the size and responsibilities of device fleets explode, the development of scalable processes to manage them is more crucial than ever. With our time and expertise spread perilously thin, advanced tooling and automation are picking up a ton of that slack. Or rather, they should be.
Even in well-resourced enterprise environments — is there such an environment? — the reality for operations and innovations teams managing device fleets is frequently far more manual and repetitive than our partners on the developer side of the house would imagine.
Too often, teams are left to manage fleets — digital signs, kiosks, point-of-sale and dedicated-purpose handsets — through a combination of tedious maintenance rounds, informal hand-raising internally and escalated customer support tickets. This approach to keeping devices online, compliant and functional — let alone keeping the content on them up to date — puts us in a permanently reactive posture, constantly fighting to triage and prioritize the most severe issues.
Most of our precious time ends up allocated to fixing problems that should not have been problems, and then implementing the needed fixes to prevent their continued (and potentially ruinous) proliferation. And we all understand it’s not sustainable. Not if fleets and the myriad business-critical functions they support are going to continue growing and innovating at this explosive rate.
Edge Device Innovation Requires Solid Management
The most visible work for any team deploying devices at fleet scale is innovation enablement — be that a hardware form factor, infrastructure integration, AI implementation or simply a new frontend software experience for customers. Executives and business leaders don’t want excuses that block these critical product rollouts. They want to get to market now.
Of course, we all understand there’s another side to that “innovate first, ask questions later” coin. When a dedicated device deployment encounters unforeseen issues and requires a fix (such as manual, device-by-device reversion to an older version of software), the work for teams managing that deployment explodes, often, far in excess of the work of the original deployment itself.
But as fleet operations experts, we can hardly say such incidents are surprising, even if the specific mechanism by which they occur may be unexpected. If we managed our fleets in a more proactive way, we could identify these “unknown unknowns” before they snowball into events that can derail entire product launches. We could be seen as invaluable collaborators instead of the clean-up crew. But how?
Meet the New Management Concept: By Exception
Managing by exception is a simple concept, at least in principle. By using a series of automated state monitors, policy enforcement mechanisms and alerts, you greatly reduce the amount of manual work and repetitive processes needed to manage a dedicated device fleet. Specifically, managing by exception lets you:
- Automate basic “check-ins” of your devices on a regular basis.
- Receive reports, scheduled or in real time, on device compliance and drift in your fleet.
- Maintain your devices’ compliance through automatic drift management.
- Focus your time on problems that cannot be remedied by automated compliance enforcement (such as self-escalation).
If you manage a device fleet, whether it’s in the hundreds or tens of thousands, truly hands-off management by exception is the Mount Everest of operational automation. And I’ll be the first to admit: We aren’t there yet. Manual intervention — even for the most sophisticated fleets — remains a reality in some situations. Machines and software fail in ways we can’t anticipate, and that means the “human touch” is still the only solution to some problems. But I’m convinced that for most of us, getting halfway up the automation mountain isn’t just achievable, it’s transformative. And that it unlocks new levels of stability, scalability and innovation.
The Future of Dedicated Device Management
In the not-too-distant future, we’ll have truly self-healing edge devices. Devices that know not just when they’re offline, but if the devices around them are offline too — and that respond to the specific circumstance in the best way (such as toggling airplane mode on and off versus trying to alert a network-connected resource of a broader outage scenario).
That future, even if it’s not quite here yet, is the one Esper was designed to support — the fully automated “manage by exception” edge device fleet. But in the here and now, whether you’re deploying content changes, AI models, firmware updates or security policies, our innovative Blueprints and Pipelines are game-changers for fleet automation. We invite you to try them, because Esper is the device management platform for operations, engineering and development teams building innovative experiences and pushing the envelope of fleet automation.
And if you want my take on how to read your organization’s device management, check out this free resource: a practical guide to “Preparing Edge Device Fleets for the Future.”
The post You Need to Stop Managing Your Edge Devices appeared first on The New Stack.
Comments are closed.