This is part of Solutions Review’s Premium Content Series, a collection of contributed columns written by industry experts in maturing software categories. In this submission, Zadara‘s Vice President of Worldwide Solution Architecture Noam Shendar offers an edge computing definition and potential roadmap to consider for the future.
We’re at the beginning of a new technology epoch.
The world has been migrating to the cloud – private, public, and hybrid – for nearly 20 years. In the last technological epoch, enterprises have utilized the cloud as a centralized repository where they can economically store and access their data. The cloud has taken over the world, replacing onsite servers and storage, and providing economies of scale. And yet, a fatal flaw with the cloud paradigm lurks.
For years, organizations have moved their data to the cloud while assuming “the cloud is the cloud.” In reality, the cloud is an earth-bound infrastructure, managed in locations around the globe, and made accessible by the internet. Accessible, however, is not the same as ubiquitous. When migrating to the cloud, considerations must be made for which data to send there, the distance between the cloud locations and the users that access them, and technologies that can mitigate some latency along the way.
The cloud tidal wave of the past 20 years has had enterprises believing that ‘the cloud will handle it’ without addressing the physical distance between cloud data centers and users, and the potential for performance, compliance, and productivity problems not only between the clouds but also between users and the cloud. These are some of the hitches that still and likely will always need to be addressed in some form with the cloud.
Fundamentally, when an organization’s data resides in the cloud, it gives up its independence for convenience and affordability. The organization’s data becomes a tenant, usually one of very many in one big cloud. A cloud that can and will occasionally fail, taking every organization’s enterprise data offline with it.
Recognizing that the cloud is a good start but not the final destination, the pendulum is swinging. We are here, on the precipice of the next technological epoch. Organizations are now shifting their focus from a centralized cloud, located out there somewhere, to a decentralized model – edge cloud. Edge cloud is the distributed computing and storage paradigm that includes the deployment of infrastructure and applications outside of centralized data centers, in multiple locations, as close as possible to where data is generated and consumed.
The very nature of edge cloud brings immediate business benefits:
- Reduced latency: Needed data is always physically close to the user
- Reduced bandwidth: More data processed locally reduces the need for it to travel over distance, sometimes dramatically
- Reduced operational costs: Reductions in bandwidth use reduce transmission costs
- Simplified scalability: The number of edge locations is increasing, and they can be onboarded quickly providing for more widespread access to data
- Security and reliability: With data stored in multiple edge locations, any data breach or system failure is less likely to be catastrophic
- More customization and control: Place and move data to where it is needed, when it is needed, with the freedom to set policies locally.
Edge cloud is also quickly becoming the key component in corporate initiatives surrounding the growth of technologies such as 5G, AI/ML, and IoT. Edge architecture is the differentiator as it supports the rapid movement and use of that data – connecting devices, making millisecond-scale decisions, and meeting the critical requirements these technologies demand – and which traditional cloud architectures cannot support.
With edge cloud, organizations can expand their presence to where they need to be, becoming the landlord of their infrastructure rather than a tenant in someone else’s. The easy integration of applications and the simplification of daily infrastructure management with edge cloud provides manifold benefits and creates an optimized environment for distributed infrastructure, instead of the one-size-fits-all cloud that has been the incumbent go-to model until recently.