AI/ML

Storage news ticker – 28 November 2025

Published

Datadobi has unveiled Advanced Storage Optimizer, which is available in the latest release of StorageMAP 7.4. It is designed to provide new levels of visibility into cost-reduction opportunities by identifying suboptimal use of storage tiers and enabling users to model scenarios before migrations. It also identifies ageing and unused data, suggesting options to reduce costs, including what to archive, along with reports that can be shared with stakeholders. Removing irrelevant data before a migration and continuously monitoring storage environments can help systems remain efficient and cost-effective over time. The latest release of StorageMAP also introduces enhanced support for Microsoft Azure Blob archiving and new automated reporting features that simplify operations and integrate with enterprise systems.

Dell has enhanced PowerProtect with Dell NativeEdge integration, Nutanix AHV protection, an Anomaly Detection review analysis page, Archive to Object enhancements, availability of the PowerProtect Data Manager Appliance, support for Azure Blob Storage, Advanced Ransomware Add-on for GovCloud and hybrid workloads, and more. The long list of enhancements can be found in a Dell blog.

Data management supplier Denodo has achieved the Databricks Validated Partner designation.

IBM says it has tripled maximum capacity for the Storage Scale System 6000 by adding support for industry-standard QLC flash storage in 30TB, 60TB and 122TB SSD configurations. A Storage Scale System 7.0.0 software update has enhancements to help organizations optimize data management, feed applications with data faster, enhance their data protection, and also offers AI support through Nvidia. Additionally, IBM introduced the Scale System All-Flash Expansion Enclosure, optimized for high performance AI training, data inferencing, HPC, and data intensive workloads. With the introduction of 122TB QLC NVMe SSDs, the Storage Scale System 6000 delivers more than 47PB of flash capacity in a single 42U rack with the All-Flash Expansion Enclosure.

Gartner has recognised Salesforce-owned Informatica as a Leader in the new 2025 Gartner Magic Quadrant Metadata Management Solutions report. Gartner says “metadata management solutions are undergoing a significant evolution, shifting from augmented data catalogs to metadata ‘anywhere’ orchestration platforms, acting as technology enablers. D&A leaders must consider the market evolution and vendor landscape as they start data and AI journeys. By 2027, organizations that actively leverage metadata analytics results across their full data management environment will reduce the time to deliver new data assets by up to 70 percent.” Get a metadata management MQ copy here.

SaaS data protector Keepit has completed the Trusted Information Security Assessment Exchange (TISAX) audit. TISAX (Trusted Information Security Assessment Exchange) is a European standard for information security assessments developed by the German Association of the Automotive Industry (VDA). It is based on ISO/IEC 27001 and evaluates how effectively organizations protect sensitive information, including customer data and intellectual property.The TISAX assessment audited Keepit’s Information Security Management System (ISMS) across a wide range of areas — including access controls, data encryption, data retention, physical security, personnel training, incident response, vendor management, and business continuity.

Prime Data Centers announced a strategic collaboration with Lambda to deploy high-density NVIDIA GPU infrastructure at their flagship LAX01 AI-ready datacenter campus in Vernon, California. The partnership combines Prime’s next-generation datacenter design with Lambda’s AI cloud platform to provide the advanced infrastructure that enterprises, research institutions, and AI-native companies need to train and deploy increasingly complex models quickly and at super-intelligence scale. Lambda told us: “We have multiple storage options that we tune for specific workload needs. Our storage is primarily targeted for hot data rather than cold storage, so all solutions are NVMe-based. We support file and object storage, and ingress and egress are enabled via DIA connections to our data center. We can also support cloud onramps or direct connections to a customer’s backbone network. We can scale these links to multi-terabit-per-second speeds.”

Lightbits Labs has partnered with Arctera to deliver SAN-class performance and high availability for Red Hat OpenShift environments, using standard Ethernet. The setup integrates Lightbits’ software-defined NVMe/TCP storage with Arctera InfoScale to provide application-level resilience, zero RPO protection, and AI-ready performance for VMs, containers, and hybrid clouds. It’s designed for enterprises moving off VMware and looking to consolidate workloads on OpenShift without sacrificing uptime or speed.

Micron announced shipping of qualification samples of its automotive UFS 4.1 product to customers. Delivering twice the bandwidth of its predecessor at 4.2 gigabytes per second (GB/s), it accelerates data access for AI models, enriching the in-cabin experience by powering features such as voice assistants, personalized infotainment and advanced safety alerts. This bandwidth in advanced driver assistance systems (ADAS) and autonomous vehicles also enables rich data capture from cameras, lidar and radar sensors to upload and feed AI model retraining and enhancement in the data center.

Nvidia has published a blog about AI agents and its AI Data Platform. It argues the platform is needed to get enterprise-held data ready for use by AI agents.

Panmnesia announced sample availability of PCIe 6.0/CXL 3.2 Fabric Switch. It’s a hybrid offering that supports both PCIe Gen 6 and CXL 3.2 protocols on a single chip, while maintaining full backward compatibility with all previous PCIe and CXL generations. Fully implemented in strict accordance with the CXL 3.2/PCIe 6.0 specifications, it ensures complete standards compliance and interoperability across all defined subprotocols. The switch can also operate in PBR mode and hierarchy-based routing (HBR) mode. Panmnesia’s PCIe 6.0/CXL 3.2 Fabric Switch Silicon is out now and early access partners can request samples and pilot systems.

A Pure Storage blog has honed in on migration to Azure with the Pure Storage Cloud Azure Native service as the destination.

Korea’s EE World has highlighted a rumor that Sandisk is in talks with Taiwan’s Powerchip Semiconductor (PSMC) to manufacture its NAND Flash memory chips amid tight supplies. PSMC will begin production in the 1st half 2026 at its Tongluo, Taiwan fab. SanDisk would supply equipment.

Korea’s 7-Eleven convenience store business and SK hynix are hooking up to sell HBM-inspired “Seven Select Honey Banana-flavored HBM Chips” snack with promotional stickers and prizes. HBM Chips is an acronym for “Honey (Honey) Banana (Banana) Mat (Flavor) Chips (Snack)”. Remember that please 🙂

According to a Hankyung report, SK hynix plans to raise DRAM production by 7x by the end of 2026. 1c DRAM production at its Icheon campus, is expected to increase by 140,000 units/month from its current at 20,000 units/month. This output will be used for GDDR7 and SOCAMM memory product, destined for AI servers.

Top 10 TrendForce 2026 technology predictions:

  1. AI chip competition intensifies as liquid cooling gains widespread adoption in datacenters
  2. Breaking bandwidth barriers: HBM and optical communications redefine AI cluster architectures
  3. NAND flash suppliers advance AI storage solutions to accelerate inference and reduce costs
  4. Energy storage systems emerge as the power core of AI datacenters and are set for explosive growth
  5. AI datacenters transition to 800V HVDC architecture, driving demand for third-generation semiconductors
  6. Next-generation semiconductor race: 2nm GAAFET production and 2.5D/3D heterogeneous integration lead the next breakthrough
  7. Humanoid robot shipments to surge over 700% in 2026, driven by AI adaptivity and scenario-based applications
  8. OLED enters new phase: premiumization of notebook displays and the rise of foldable smartphones
  9. Meta accelerates the global advancement of near-eye displays as LEDoS builds momentum
  10. Autonomous driving accelerates: passenger cars standardize assisted driving while robotaxi expands globally

TrendForce’s latest research shows that significant increases in conventional DRAM contract prices, higher bit shipments, and growing HBM volumes drove the global DRAM industry revenue to US$41.4 billion in 3Q25, marking a strong 30.9 percent QoQ growth. DRAM suppliers’ inventories are almost depleted, and growth in bit shipments will decrease substantially. CSPs continue to be relatively flexible with procurement costs, which causes other applications to also increase their prices to secure supply. Contract prices for both advanced and legacy nodes, as well as across all major applications, are anticipated to increase rapidly. TrendForce predicts that conventional DRAM contract prices will rise by 45–50 percent QoQ, and total contract prices (including DRAM and HBM) will increase by 50–55 percent.

VAST Data announced a strategic collaboration with Lablup Inc, developer of the open, scalable AI computing platform Backend.AI and a participant in South Korea’s government-backed Sovereign AI Project. The companies have built a high-performance, scalable data foundation that enables Korea’s national AI consortium, which aims for technological self-reliance, to train and deploy large-language models entirely on domestic infrastructure. Backend.AI provides a unified environment for AI model training, deployment, and inference across GPU and CPU resources. By integrating the VAST AI Operating System, Lablup and its customers can manage training data, model checkpoints, and artifacts with unmatched throughput and resilience, ensuring smooth data flows from model development to real-time serving.

Virtuozzo has put together a benchmark datasheet that compares Virtuozzo Storage vs. Ceph on real-world workloads, focusing on:

  • Backup and restore at scale
  • Hot VM/application performance
  • Usable capacity with erasure coding
  • Operational effort and TCO

Cloud storage provider Wasabi Technologies is updating its EMEA Partner Programme and actively recruiting partners in the Nordic, Benelux, and Southern Europe regions. The new Wasabi Systems Integrator Programme gives partners the technical tools and commercial enablement they need to deliver solutions spanning AI, physical security, and cyber resilience.

Testing by SW RAID supplier Xinnor demonstrated that xiRAID Opus transforms Intel Xeon 6 E-core processors into viable solutions for high-performance storage, a task previously out of their league. Using just 8 cores, we’re told the solution achieved 96 percent efficiency (1.1M IOPS) for RAID6 random writes and 94 percent efficiency (42.3 GB/s) for RAID6 sequential writes, while maintaining native sub-10μs latencies. For RAID10 configurations, E-cores delivered 100 percent efficiency (1.79 M IOPS) on random writes and 83 percent (25.1 GB/s) on sequential writes with the same core count. Read more in Solution Brief document here.

XDA Developers have shown that non-powered SSDs slowly lose data (bit rot) as trapped electrons in a floating gate cell tunnel through the insulating barrier over time. Details here. Don’t use SSDs for archival storage unless you power them on every 6 – 12 months and have the controller automatically refresh decaying cell contents.

Nigeria-based wholesale open access digital service provider Backbone Connectivity Network (BCN) and Zadara partnered to launch Africa’s first Neutral Multi-Tenant AI Factory powered by Nvidia GPUs. This will enable cloud providers across Africa to provision AI-ready clouds. The Zadara-powered Neutral AI Factory, launching in Nigeria, helps cloud providers overcome high upfront costs. As one of the first NVIDIA design and integration partners to support multi-tenant architectures, Zadara enables BCN to deliver flexible, on-demand AI clouds.