IT must adapt to the shift of computing and storage to the edge.

• The growth of edge computing is a significant indication that IT is expanding beyond the data center.
• The distributed computing model, known as Edge, brings functions such as computation, data analysis, and storage closer to the sources of the data.
• By reducing latency, applications can run faster and provide faster access for users compared to applications hosted on servers in a data center.

The emergence of edge computing, which involves processing data near its source, is a clear indication that IT is expanding beyond the data center.

By using edge technology, users can physically be closer to servers, which can improve application performance and reduce latency compared to servers located in a data center.

The International Data Corporation forecasts that worldwide spending on edge computing will increase by 15% from $2023 to $232 billion in 2024. By 2027, combined enterprise and service provider spending on edge computing hardware, software, professional services, and provisioned services will reach nearly $350 billion.

Gartner's Thomas Bittman stated that companies typically employ edge computing in three specific situations.

When they require a quicker reaction than a far-off data center can offer, when it's more cost-effective to process data near the user rather than sending it elsewhere, or when they lack a dependable network connection to a distant data center.

Bittman stated that Edge can offer both top-line benefits, such as new business opportunities, and bottom-line benefits, including increased automation, efficiency, and productivity.

Bittman stated that edge computing complements cloud computing by extending digital transformation to the edge, allowing companies to capture more data, gain insights more quickly, augment physical interactions with digital information and controls, and put intelligence where it needs to be.

According to Michele Pelino, principal analyst at Forrester Research, low latency and real-time analytics are necessary for specific use cases such as autonomous vehicles and real-time predictive analytics. This can be achieved through edge computing, which allows data to be processed locally before going to the cloud for additional analysis using artificial intelligence.

Gartner research indicates that edge computing is valuable in all industries, although its application varies widely. According to Bittman, edge computing may be useful in enterprises with widely distributed people or assets, where interactions occur at the edge, such as shopping, assembling, collaborating, and utilizing various systems.

According to Bittman, the sectors that are most distributed or asset-heavy will use edge computing the most, including manufacturing, energy, retail, transportation, and defense.

Real-time monitoring of machine health is a key use case of edge computing in manufacturing, as it allows for predictive maintenance to analyze and detect changes in equipment, machinery, or operational processes before a failure occurs, Pelino stated.

Pelino mentioned that in addition to optimizing production processes, other edge use cases in manufacturing involve utilizing robotics, drones, or autonomous guided vehicles to perform critical industrial operations.

Sectors that can benefit

The most data is generated at the edge for manufacturing and energy, which requires automation systems. Energy also has a volatile power grid and requires rapid reaction and predictive responses. Retail is highly distributed and requires agile local updates on pricing, applications, sales, and automation.

Edge computing can benefit the financial services sector, particularly high-frequency trading and hedge fund firms, by reducing latency through the deployment of on-premises edge servers close to the trading floor, minimizing latency or last-mile data delays between servers located at different exchanges. This results in maximum profits on high-volume, low-margin trades.

Data protection is a significant obstacle to edge computing deployment, according to Pelino. The complexity of managing security across a variety of edge devices, use cases, and environments makes it challenging for companies to address this issue. This challenge is compounded by the fact that attackers can exploit standard application flaws, network misconfigurations, insecure protocols, or physical access to gain access to edge environments.

Hackers can exploit every new network connection, smart device, edge server, or micro data center, according to Pelino.

To address security concerns, Pelino recommended implementing a zero-trust edge architecture that integrates networking and security, starting with on-premises software defined wide area networks, firewalls, and strict network access, and ending with routing, secure web gateways, and cloud security gateways.

Bittman stated that smartphones have created security challenges by extending the digital perimeter, and that edge computing is like the smartphone security challenge on steroids.

To ensure all connections are secure, it is necessary to encrypt data on devices, protect the backhaul to cloud, and monitor the behavior of edge devices, according to Bittman.

Bittman pointed out that expanding the digital perimeter beyond the data center will necessitate the implementation of new cybersecurity techniques and AI for behavior monitoring.

Beyond security, companies can face other barriers to adoption of edge technologies.

"The fragmented vendor market for edge computing presents a diverse set of technologies and solutions that, when managed together, address key business scenarios, workloads, and use cases," Pelino said.

"Vendor-to-vendor and vendor-to-enterprise partnerships are essential for creating a comprehensive edge solution due to the diverse nature of the edge landscape, according to Pelino."

A lack of collaboration among IT and operational technology stakeholders can be another potential barrier.

"Tech executives must break down the IT and OT silos that often exist among the many owners of edge use cases, Pelino said."