IBM Goes Container-Native With Spectrum Fusion

By Patrick Moorhead - May 14, 2021

The realities of deploying cloud-native workloads force a different way of thinking about storage. Data that is location-independent introduce a flurry of new challenges. Cloud-native container technologies place different demands on persistent storage. Storage software must evolve to address these challenges.  

Data management and data placement are complex problems to solve. Ensuring consistent access and availability across a hybrid-cloud infrastructure is a challenge. Achieving the proper levels of cyber-resiliency becomes incredibly difficult. All of this must be addressed while maintaining a level of performance that doesn't stall the business-critical applications dependent upon that data. 

The same is true for the realities of designing storage for AI-enabled workloads.  AI is now a mainstream enterprise workload. Keeping the powerful compute workhorses supplied with data requires storage designed for AI.  

These realities are driving the thinking behind IBM’s latest storage announcements. Let’s delve into what IBM is delivering. 

A New Container-Native Storage Foundation 

Storage architectures are evolving to meet the needs of cloud-native workloads. The industry is finding that storage needs to go beyond simply delivering CSI drivers for container integration.  Storage services must become container-native to meet the needs of cloud-native workloads. 

Container-native storage is a trend that we see play out in real-time, and IBM is front-and-center in driving this trend forward. IBM's newly announced Spectrum Fusion is the company's answer for delivering storage solutions into a cloud-native world. 

IBM Spectrum Fusion delivers the elements critical to addressing the pain-points encountered when deploying cloud-native architectures into the enterprise. Spectrum Fusion brings together the unstructured data management capabilities of IBM Spectrum Scale, the metadata management capabilities found in IBM Spectrum Discover, and the data protection elements of IBM Spectrum Data Protection Plus.  

IBM will deliver these capabilities first as a new HCI solution, IBM Spectrum Fusion HCI.  Spectrum Fusion HCI combines the cloud-native goodness of IBM’s Spectrum software suite with Red Hat's OpenShift container management software, delivering an integrated HCI appliance for both containers and virtual machines. 

The base configuration for the IBM Spectrum Fusion HCI includes a 42U rack, two 100GbE ethernet switches, two ethernet management switches, and six 1U x86 storage/compute nodes. Compute can scale to up to twenty nodes, while storage scales with up to ten 7.68TB NVMe flash drives per node.  That's a lot horsepower. 

Spectrum Fusion HCI delivers the unified and straightforward management experience that the industry has come to expect from HCI solutions while also providing a unique set of cloud-native storage capabilities powered by proven IBM storage software.  The solution is compelling. 

IBM Spectrum Fusion HCI will ship later this year, while a non-HCI version will be available sometime next year. 

Storage for AI 

Container-native isn’t the only disruptive technology invading the data center. AI and machine learning are rapidly becoming mainstream in the enterprise.  Much like the adoption of containers, AI requires thinking differently about storage.  This is an area where IBM has long led with its highly performant storage arrays. 

IBM today announced an update to its IBM Elastic Storage System. IBM's ESS 3200 is a welcome update, doubling the performance of its previous generation ESS 3000.   The ESS 3200 delivers up to 80GB/s per node pair.  The ESS 3200 delivers nearly twice the performance advertised by IBM's nearest competitor in this space. 

Beyond the compelling performance promised by the IBM ESS 3200, there are also some nice software updates.  The new platform includes flash wear monitoring with call-home and alert capabilities, remote call-home diagnostics, and software upgrades that can be automated with updated Red Hat Ansible automation and configuration. 

IBM has also updated its ESS 5000 product with new storage capacity improvements. The update allows for up to 380TB/rack unit for the ESS 5000 SC models while supporting up to 264 TB/rack unit for the SL models.  This is a nice capacity bump for the product.

The Analyst’s Take

Data is a competitive differentiator. Applications and data must adapt to the demands of the business. It's all about agility.  Agility is driven by flexible software. 

IBM delivers some of the most compelling storage solutions on the market. The performance and reliability provided by IBM's storage arrays cannot be understated. At the same time, the secret weapon in IBM's arsenal is its portfolio of storage and data management software. 

IBM's various Spectrum software offerings allow the company to respond very quickly to shifting trends in IT architecture.  This was true as the industry embraced hybrid-cloud architectures. It continued to be true as AI became commonplace. And it's true today, as IBM addresses the needs of cloud-native workloads with container-native storage solutions.  The addition of Red Hat's OpenShift and Ansible technologies only sweetens the pie. 

Focusing on IBM's storage hardware misses the point.  The innovation in storage today is all software-driven, and software-driven solutions are what IBM delivers best.  

Note: Moor Insights & Strategy writers and editors may have contributed to this article. 

+ posts

Patrick founded the firm based on his real-world world technology experiences with the understanding of what he wasn’t getting from analysts and consultants. Ten years later, Patrick is ranked #1 among technology industry analysts in terms of “power” (ARInsights)  in “press citations” (Apollo Research). Moorhead is a contributor at Forbes and frequently appears on CNBC. He is a broad-based analyst covering a wide variety of topics including the cloud, enterprise SaaS, collaboration, client computing, and semiconductors. He has 30 years of experience including 15 years of executive experience at high tech companies (NCR, AT&T, Compaq, now HP, and AMD) leading strategy, product management, product marketing, and corporate marketing, including three industry board appointments.