Cloud computing was one of the defining technologies of the 2010s. Companies increased their investments in cloud-based compute, storage and networking resources to run applications and store data with greater flexibility and scalability than ever before.
But now with the growth of hybrid cloud architectures and the rapid evolution of artificial intelligence (AI) and machine learning (ML), the edge - and how it connects to the cloud and to the data center - has become a more central concern.
What is the edge (and what is edge computing)?
The edge is a hardware location at the periphery of a network, where data is collected; it exists outside of a data center. That data may be passed on to a data center or cloud without modification, or subjected to further processing before going there.
Red Hat has made a distinction between “the edge” and “edge computing,” with the latter referring to the act of processing data before transmitting it anywhere else. The edge itself consists of devices that store data.
Why is the edge important?
Although cloud environments provide vast resources at economical cost, they aren’t ideal for the most performance-sensitive use cases. These include sending huge amounts of data needed for training an AI model, or running a workload that pushes a graphics processing unit (GPU) to its limit.
In these instances, the sheer physical distance between the cloud and where the data is collected means that there will be noticeable latency. The volume of information being transmitted may also create bottlenecks with bandwidth.
For example, AI algorithms must be fed a lot of labeled data before they can function properly. Collecting, storing and processing that data at the edge will allow for quicker, more economical AI training than trying to do the same in a faraway cloud.
Or consider the case of a hybrid cloud. An edge storage solution could provide a critical piece in connecting the data center/private cloud to a public cloud, helping to manage the flows of large quantities of data between diverse environments.
Overall, the right hardware and software on the edge help:
- Make data more readily accessible and available for demanding use cases.
- Provide top-notch performance that minimizes latency.
- Support the growth of AI and hybrid cloud computing initiatives.
IDC has predicted that up to 50% of all enterprise IT infrastructure will be deployed at the edge by 2023, compared to just 10% in 2019. Meanwhile, according to Grandview Research, the global AI market is expected to expand at a 40% compound annual growth rate from 2021 to 2028, underscoring the need for resilient edge solutions.
Enter IBM Spectrum Fusion and Elastic Storage System (ESS) 3200.
What are IBM Spectrum Fusion and ESS 3200?
IBM, a pioneer in hybrid cloud computing, has a broad range of edge offerings, designed to support better data accessibility and availability, with the two below being among the most prominent:
IBM Spectrum Fusion
This is a container-native hybrid cloud storage platform, built for Red Hat OpenShift. It enables:
- Straightforward global access to data, from any resource.
- Resiliency and scalability for mission-critical workloads.
- Faster business insights, e.g. from AI algorithms.
- Consistency across hybrid cloud environments, thanks to containerization.
- EAsier development of cloud-native applications.
IBM Spectrum Fusion’s first incarnation is in a hyperconverged infrastructure (HCI) configuration known as IBM Spectrum Fusion HCI, according to Steve McDowell of Moor Insights and Strategy, writing for Forbes. This offering combines high-performance hardware and software to support GPU-accelerated applications, hybrid cloud integration and unified management of containers and virtual machines, among other capabilities.
The ESS 3200 is a software-defined edge storage appliance built for AI and hybrid cloud modernization. Compatible with all ESS models, the ESS 3200 combines super-fast NVMe storage with containerized software and a compact 2U enclosure.
This solution helps accelerate the time to value for AI and other projects, thanks to high throughput for, and smart management of, data stored on the edge. It may be deployed by itself or in combination with other ESS hardware.
Join us for a Spectrum Fusion and ESS 3200 webinar
InfoSystems is hosting a July 8, 2021 webinar that will dive into more detail on both of these solutions within the context of the edge. Register now or connect with the InfoSystems team for more information.
For over 25 years, InfoSystems has provided reliable IT solutions to build and maintain strong and secure systems for both SMB and enterprise organizations. Headquartered in Chattanooga, TN, our trusted team of experts specialize in traditional infrastructure, IT optimization and cybersecurity services, as well as next gen solutions such as hybrid cloud and artificial intelligence, from partners such as IBM, Dell Technologies, Red Hat, VMware and Cisco.