Did you miss a session at the Data Summit? Watch On-Demand Here.
It comes down to simple physics and cost: In cloud computing, high availability and sub-second response times are nearly impossible, or at least unfeasibly expensive.
This drawback has given rise to edge computing, which moves computing resources to the physical location of data creation, or the so-called “edge” of the internet. The touted results are real-time speeds and dramatically increased availability, flexibility, resiliency and consistency of the data.
According to Dave McCarthy, research vice president for cloud and edge infrastructure services at IDC, there has been a shift in mindset from “…anything and everything should go to the cloud, to ‘Let’s use the cloud for what it’s good for, and use other things when they make more sense.’”
Edge computing has increasingly become a priority for a growing number of organizations. According to IDC, worldwide enterprise and service provider spending on edge hardware, software and serves is expected to hit $176 billion in 2022, representing a 14.8% increase over 2021. That spend is anticipated to approach $274 billion by 2025, according to the firm. Similarly, the LF Edge arm of the Linux Foundation expects edge spending to jump to $800 billion by 2028.
This has equaled exponential growth in the number of providers. Industry giants and specialized companies alike have been expanding into the space; established providers of edge computing platforms and services include Cloudflare, Macrometa, Platform9 and Litmus Edge. Amazon Web Services (AWS) offers its [email protected] technology, while IBM has Watson Anywhere, and nearly all other IT vendors, from Google, to Dell, to Hewlett-Packard Enterprise (HPE) have announced plans to provide some sort of multiple edge computing platform.
Couchbase, a distributed NoSQL cloud database, has also expanded its support of edge computing with the launch of Couchbase Mobile 3. The new platform allows developers to build fully native applications in the cloud, at the edge and on mobile and IoT devices using their chosen languages, frameworks and platforms, according to Wayne Carter, Couchbase’s vice president of engineering.
Better, faster, stronger — the demand for data and apps
The 11-year-old, Santa Clara-based public company has established itself with its two versions of an open-source, NoSQL, multimodel, document-oriented database software.
As Carter noted, modern apps need to be faster, more resilient, agile, and accessible, and have the capability to be run from anywhere. Since apps run on multiple different systems, developers need to be able to configure hundreds of locations and devices quickly and easily.
“Customers increasingly require mobile and edge capabilities to meet modern application demands, and data must always be available, so apps perform at unmatched speed,” Carter said.
The goal with the Couchbase platform, and with edge in general, is to move data closer to where it is being used, even when it is in motion, to ensure that apps always have access to it.
One of the reasons edge computing has become so popular is it helps address use cases that the cloud can’t, IDC’s McCarthy explained. Using the cloud is cost-effective and fast, but it has performance-related limitations. Apps that rely solely on centralized cloud data centers to process and store data are subject to latency and downtime whenever internet connectivity is slow or frequently interrupted. The time it takes to send a command to the cloud, have the cloud process it and send the information can be prohibitive. What’s more, top cloud providers have had significant outages lately.
“How can you continue to operate if the cloud isn’t available, or the network between you isn’t available?” McCarthy posited.
Couchbase 3 addresses this gap by providing sub millisecond response times, Carter said. Data integrity is maintained with automatic synchronization across edge and mobile infrastructure, regardless of internet connectivity.
Developers can use the platform in edge data centers, in the cloud, on 5G networks, on-premises or edge devices. This multi-tiered, hierarchical architecture support allows it to meet any speed, availability, technical or security requirements, Carter said. The result is that apps are fast and resilient and not dependent on, or impacted by, distant cloud data centers or variability in network speeds. Apps can also be developed and deployed in a way that meets increasingly restrictive governance and security requirements.
Device ubiquity means that if a device is on, data is available, and it is always synchronizing, Carter said. Given the laws of physics, you can only achieve a certain speed with cloud computing. “You can’t solve this problem any other way,” he said.
As an example, the platform has been used by a leading airline to digitize its pre-flight check process. It has been embedded onto tables for recording inspections, synchronizing data across crew tables in real-time, even when those devices are disconnected. This has improved accuracy and safety while ensuring on-time departure, Carter explained. In similar cases, the platform has underlaid airplane meal ordering systems so that all flight attendants have visibility into inventory (say, the number of available turkey sandwiches or cans of ginger ale).
Fostering creativity on the edge
Couchbase 3 is certified on Amazon Web Systems (AWS), Verizon, Google Cloud Platform (GCP) and Azure. Reference architectures and deployments have been developed for AWS Local Zones, AWS Wavelength, AWS Outposts and Verizon 5G Edge.
“What Couchbase has done is taken their success in cloud databases and extended those feature sets to smaller edge environments,” IDC’s McCarthy said. “It enables this world of apps that can span from the cloud to the edge.”
Due to its vastness, McCarty acknowledged that the edge is a concept that can be difficult to understand. “If it’s so new, how can it be so big?” he said.
He described it as a shift from a centralized computing model to a more distributed computing model and a market driven by the expansion of IoT and AI applications. “It’s a big market partially because existing things are being modernized,” he said. “The edge makes up a lot of different deployment scenarios.”
Edge computing also fosters creativity, McCarthy said, as it enables developers to create more tools and app features and options. “You can get all the benefits of cloud without being restricted to just being in the cloud,” he said. “It’s a sort of best of both worlds’ scenario.”