Nov 30 2020

Edge Analytics vs. Cloud Analytics: Which Is Right for You?

Cloud analytics rules the enterprise these days, but interest in edge analytics solutions is growing. Luckily for businesses, they might not have to choose.

The cloud is continuing to gain steam as businesses brace for long-term remote work. But when it comes to data analytics, we may see the pendulum swing toward the edge.

Edge analytics, the concept of analyzing data in real time in a localized form, has been growing more popular. However, many organizations remain unsure about it, especially given the wide array of options.

In many ways, edge analytics is still a technology finding its way to maturity, unlike cloud analytics, but they could find a path together. Here’s how.

What Is Edge Analytics?

Edge analytics is a natural extension of edge computing, which puts computing resources close to where they are needed rather than accessing them remotely.

Edge computing reflects a movement of computing concepts away from a centralized medium and into a decentralized area. Abhijit Sunil, an analyst with Forrester, notes that this reflects a gradual historic shift over a number of decades as networking and technical capabilities improved.

“At first, there were these big mainframe machines, then it got disaggregated,” he says. “Then, it came back together in one place: the cloud. Now, it is going back out in the form of edge computing.”

Edge analytics uses these computing resources to help conduct complex analysis of data locally, with an emphasis on real-time use cases — an area expected to grow by nearly 30 percent by 2025.

What Are the Benefits of Edge Analytics?

The strongest benefit of edge analytics is the benefit of improved latency; data can be analyzed and used both instantly and locally.

One use case is for autonomous vehicles, which use edge analytics in their operations. Computational tasks often must happen in a matter of seconds, which means that the analytical work must be handled instantaneously. That could mean cloud computing would be too slow to react in time. Such use cases, says Sunil, will benefit from the rollout of 5G technology.

Additionally, as IBM notes in an explainer on its website, processing analytics at the edge can help minimize security concerns: Data travels shorter distances, which creates less risk of disconnection and allows for continuous operation.

Contexts for edge analytics are wide. Smart cities, for example, can use edge analytics to assist with improving traffic safety on the fly.

Sunil points to specific edge analytics use cases, such as real-time analysis for oil and gas companies, as well as to broader industry implications such as in-store tracking in retail contexts, or personalized advertising.

“Big retail chains would like to know how somebody’s browsing online,” he says, “and then from there, the customer’s behaviors inside the store.”

Edge analytics can also empower occupancy management for social distancing — a major issue as offices and factories attempt to reopen safely amid the COVID-19 pandemic.

MORE FROM BIZTECH: Is Analytics-as-aService the Big Data solution for small businesses?

Edge Analytics Providers

Edge analytics is often a combination of hardware and software working in tandem as a single solution, and what works for one organization may not work for another.

“Edge computing is very use case-specific,” Sunil explains, noting that it may require working with multiple vendors to help fully roll out a solution that meets your needs.

“Investments in edge computing can closely follow a framework, but it will heavily depend on the use case itself,” he adds. “The framework will have parts: investment in the infrastructure, investment in working with a network player and investment in hosting or services close to where you are. It could be a public cloud; it could be the data center.”

Ultimately, technology solutions will roll out across disciplines, requiring the use of Internet of Things solution providers, tools to build out data center solutions in locations where they’re needed and business intelligence software from vendors such as IBM and Tableau.

The cloud could also play a key role in maximizing the edge, particularly in the case of content delivery networks (CDNs), which have seen traffic increases of 30 percent or more in the months since the COVID-19 pandemic broke.

“The biggest players in cloud computing at this time, like the major public cloud players, and the enablers of cloud computing, like the CDNs and the network players, all of them will have a play in the edge computing space as well,” Sunil says.

What Is Cloud Analytics?

There are several benefits to edge analytics, but cloud analytics still serves businesses as well. Cloud analytics uses cloud resources to analyze different types of data. Use cases can vary significantly — from real-time reporting to in-depth Big Data solutions.

Despite the field’s maturity, there is still plenty of room to grow into cloud-based analytics. Research from IDC finds that the market for data analytics will grow to $247 billion in 2022 — a nearly 25 percent increase from this year’s projection.

Trying to decide between cloud analytics or edge computing isn’t a case of choosing one over the other. Often, they work together, especially during processes that are connected to the physical world rather than wholly online experiences.

What Are the Benefits of Cloud Analytics?

One of the strongest benefits of cloud analytics, especially compared to edge analytics, is that it can be compartmentalized and managed remotely using tools such as Infrastructure as Code (IaC). This means that it can be easier to scale than edge analytics. The compartmentalization of the cloud and the ability to dole out resources on the fly is a major strength.

A cloud service can also handle lots of complex information at a moment’s notice. Discussing the use of remote cameras, IBM Executive Cloud Architect Ashok Iyengar and Senior IT Architect and Data Scientist Ivan Portilla explain in a blog post that cloud analytics excel at more in-depth types of data and offer more access to processing power. The authors note that timing and jurisdiction play key roles in deciding where to allocate analytical resources.

“Two factors dictate that answer: How critical is it to analyze the data in real-time and whether additional analysis needs to be done with that data,” they write. “Then, there’s that storage requirement (or not) to meet business and jurisdictional compliance requirements.”

Ultimately, Sunil explains, there may be room to use both.

“Edge computing has very specific use cases, which will kind of augment the way the cloud computing landscape is at this time,” Sunil says.

Cloud Analytics Providers

Major cloud platforms such as Microsoft Azure, IBM Cloud and Amazon Web Services offer a variety of solutions for managing cloud resources. Those resources include cloud analytics tools that can process a variety of analytical types, including interactive analytics, Big Data solutions and dashboards.

As a function of the cloud’s maturity, analytical options have a level of breadth and flexibility that can adapt to any use case. Sunil emphasizes that the maturity of cloud offers a sharp contrast to solutions such as edge analytics.

“Edge computing is still evolving,” he says. “Many equations, many kinds of variables are part of that edge investment.”

Getty Images/ GaudiLab