Storage News Ticker - 8 May

Data source connector Airbyte launched Airbyte Agents, a context layer that gives production-grade agents direct access to a unified, search-optimized index of an organization’s data that is replicated and ready to query before the agents run. It’s based on Context Store, a replicated, search-optimized index that unifies a company's data across systems before the agent ever runs.

For example, customer records from Salesforce, tickets from Zendesk, issues from Jira, and conversations from Slack are brought together into a single queryable index with history and state preserved. The work of assembling context happens in advance, not at query time, so agents query the Context Store instead of chasing live APIs. That typically collapses five or six calls to one or two and dramatically reduces token consumption.

Airbyte Agents is available now through the Model Context Protocol (MCP), which works inside Claude, ChatGPT, Cursor, and any MCP-compatible client, and through a native SDK for teams building custom agents from the ground up.

…

Data trust supplier Ataccama announced a partnership with ServiceNow which will see Ataccama data quality scores, findings, and observability signals directly surfaced on the ServiceNow Data Catalog. This is so users and AI agents can see whether an asset is fit for use before they rely on it.

…

Broadcom announced VMware Cloud Foundation (VCF) 9.1 with:

• Lower-cost AI infrastructure: Up to 40% reduction in server costs, up to 39% lower storage TCO, and up to 46% reduction in Kubernetes operational costs for AI workloads at scale.
• Unified platform for AI workloads: Supports inference and agentic AI alongside traditional VMs and containerized applications on a single infrastructure layer, reducing operational complexity.
• Expanded hardware flexibility: Mixed compute infrastructure support across AMD, Intel, and Nvidia, including multi-accelerator GPU options and high-speed networking capabilities for AI workloads.
• Built-in security and governance: Integrated zero-trust security, ransomware recovery, continuous compliance enforcement, and live patching to protect AI models, data, and applications.

Check out a series of VCF 9.1-related blogs from Broadcom here. 

Scandinavia-based backup and recovery system supplier Cristie is now Rubrik Black Belt certified which proves its ability to design, implement, and operate advanced recovery architectures built on Rubrik Zero Trust principles, with verified restores, automated testing, and real‑world cyber recovery readiness. It has introduced a new release: Cristie Recovery Solution – Black Edition. Read more here.

...

Immutable and cryptographically verifiable enterprise-scale database supplier Codenotary released immudb v1.11, a major update to its open source database that adds immutable audit logging and expands PostgreSQL compatibility. The only immutable enterprise-scale database with cryptographic verification, immudb ensures data cannot be secretly changed or deleted. With version 1.11, this same protection now also applies to database activity – meaning not just the data itself, but also who did what and when can be recorded in a way that cannot be tampered with.

We’re told that Immudb is used by many thousands of organizations worldwide to create solutions that require a zero-trust approach to their sensitive data. There have been more than 50 million downloads of immudb. Immudb 1.11 is open source and available on GitHub at https://github.com/codenotary/immudb.

…

Everspin announced a $40 million agreement with a U.S. prime contractor to provide Toggle MRAM process technology capabilities and engineering services for US defense programs. Everspin is the only current provider of MRAM. Under this agreement, Everspin is a subcontractor on an existing Prime Contract and will provide engineering and foundry services for the U.S. Department of War (DoW). Everspin’s recently announced foundry services agreement with Microchip expands its U.S.-based manufacturing capability, which is expected to support future production of these programs. The agreement is valued at an aggregate of $40.0 million over two and a half years.

…

Previously moribund data archiver FalconStor has started growing again with double-digit revenue and net income growth. Its Q1 2026 revenue increased 76% Y/Y to $4.3 million while income went up from the year-ago $100K loss to $1.1 million. It’s “seeing a growing number of IBM-centric MSPs standardizing on StorSafe-MSP as a core platform for hybrid cloud data protection and archive services. At the same time, our reach continues to expand across enterprise customers through StorSafe-Enterprise and our Habanero offering, extending our presence across both on-premises and cloud environments.” It said: “Our hybrid cloud annual recurring revenue (ARR) run-rate increased 89% year-over-year, and ARR represented 87% of total billings in the quarter, up from 67% in the prior year period.” 

…

Data mover Fivetran released “The 2026 Agentic AI Readiness Index,” measuring how prepared enterprise data environments are to support agentic AI workloads/initiatives in production. only 15 percent of organisations are fully prepared to support agentic AI in production, even as nearly 60 percent report investing millions to tens of millions in the technology. Check out the full findings in The 2026 Agentic AI Readiness Index here.

…

Kingston expanded its DC3000ME PCIe Gen5 U.2 NVMe SSD family with a high-capacity 30.72 TB option built with 3D eTLC NAND. It delivers sequential read speeds of up to 14 GB/s and random read performance of up to 2.8 million IOPS. The drive has on-board power loss protection (PLP) and supports AES 256-bit encryption and TCG Opal 2.0 self-encrypting drive (SED) capabilities. For more information visit kingston.com. 

…

NetApp has hired Alvaro Celis as its Chief Partner and Ecosystem Officer, reporting into Chief Commercial Officer Dallas Olsen. Celis will lead the company’s World Partner Organization and deepen relationships with NetApp’s ecosystem of partners and alliances. He joins NetApp after three decades at Microsoft, where he most recently led global partner, GTM and commercial sales and brings deep experience scaling multibillion-dollar AI and cloud businesses. He oversaw a partner ecosystem that generated over $100 billion in annual revenue and helped drive over 30% YoY growth in Microsoft’s SMB business segment. 

…

Peer Software announced that McClure, a multidisciplinary, multi-office, engineering firm serving public‑ and private‑sector clients nationwide, has deployed Peer Global File Service (PeerGFS) to enhance its distributed storage infrastructure. With PeerGFS in place, McClure’s growing organisation benefits from reliable, high‑performance access to large design files, enabling faster collaboration and improved productivity across its engineering teams.

…

Enterprise software developer Proxmox announced the release of Proxmox Backup Server 4.2. The highlights are:

• Based on Debian 13.4 “Trixie” with Linux kernel 7.0 as new stable default 
• Includes ZFS 2.4 for enterprise-grade storage 
• Move groups and namespaces within a datastore 
• Server-side encryption/decryption for sync jobs 
• Improved sync performance with concurrent group processing 
• S3-compatible object stores supported as backup storage backend, with request statistics and notifications 
• Numerous performance, usability, and backend improvements 
• And much more… 

Proxmox Backup Server 4.2 is available for free download now at https://www.proxmox.com/downloads. The software is published under the GNU AGPLv3 license.

…

Scale-out filesystem supplier Qumulo has won Future's Best of Show Award at the 2026 NAB Show in Las Vegas. Qumulo's Data Platform was singled out for its ability to meet the need for infrastructure to cope with studios, broadcasters, and streaming platforms grappling with an explosion of high-resolution content and the distributed production pipelines needed to deliver it.

…

Semiconductor designer Rambus has PCIe 7.0 Switch IP with Time Division Multiplexing that:

• Supports bandwidth scaling, low latency, and efficient data movement for AI, cloud, and HPC systems
• Increases link utilization through intelligent traffic multiplexing, enabling simpler architectures and scalable disaggregated and pooled compute designs
• Extends the industry-leading Rambus PCIe IP portfolio which spans switches, controllers, retimers, and debug solutions to support next‑generation AI infrastructure 

…

Electronic Healthcare Record (EHR) supplier ShelterZoom announced SpareTire to provide zero-downtime continuity when systems are hit by cybersecurity threats. It says it establishes continuity during downtime as the new normal. Through ShelterZoom’s proprietary blend of external-cloud resilience, comprehensive downtime continuity across critical systems, cryptography-based data tokenization and trusted AI infrastructure, Spare Tire paves the way for a digital future where data, content, and infrastructure are resilient, controlled, verified, and protected.

This proactive approach enables users to continue working in real time with synchronized, validated data without interruption, data loss or delays, redefining how organizations operate through failure. Instead of recovering after disruption, organizations maintain continuous operations without downtime, shifting from a reactive model to true operational resilience.

Find out more here.

…

StorONE says customer Storage Guardian, by using StorONE’s hybrid storage, went from 9 racks to 2, cut space by 80% and power by 78%, and improved backup and DR performance. Storage Guardian supplies DRaaS and data protection services. See more in a StorONE webinar.

…

Troveo provides licensed real-world video data sets for AI and has announced an expansion of its platform into five new data categories: audio, text, agentic trajectories, gameplay and egocentric robotics. Agentic trajectories are enterprise workflow traces; real-world business data sourced directly from companies across a range of industries that captures actual enterprise workflows. Gameplay is video game data, including time-synced keystroke and character progression metadata used for frontier world models. Egocentric robotics is real-world, first-person perspective data from real operating environments that power the robotics world. 

Marty Pesis, founder and CEO of Troveo, said: “Beyond access to compute and top-tier talent, training data remains the biggest bottleneck for building the next generation of AI models. The most valuable data for solving that is real-world, meaning it captures the complexity of how people actually live and work. It is clean, accurately labeled and ready to train on. And it’s non-public, meaning it has not been incorporated into a prior training run. It lives in archives, hard drives and operating environments that nobody has indexed or packaged for AI. Troveo delivers this data directly into the training environments of the worlds’ top labs.” Learn more at troveo.ai.

…

Veeam has been ranked #1 for worldwide data protection software market share by IDC in its latest IDC Semiannual Software Tracker, 2025H2. The tracker reports that Veeam captured 13.6% market share in 2025H2, increasing from 13.2% in 2025H1. IDC also reports that Veeam delivered 11.5% sequential growth in 2025H2, outpacing the market average of 8.8%.

…

Veeam has released a Q1 ransomware report which reveals:

• AI shrinks the patch window to hours, not days: Agentic AI is accelerating vulnerability discovery and exploit development, increasing the risk of rapid, large-scale attacks.
• Fewer organisations are paying ransoms, but the value of payouts is rising: Payment rates remain relatively low at 23%, but the average ransom payment has increased by 15% to $680k 
• Data-theft-only extortion is losing effectiveness: Organisations are refusing to pay for exfiltration alone, as it rarely reduces long-term risk.
• Identity-based attacks dominate: Most breaches now stem from compromised logins (SSO, SaaS, MFA resets), with attacks often appearing as legitimate access.
• Mid-sized firms and critical sectors most at risk: Companies with 11–1,000 employees account for the majority of attacks, with healthcare and consumer services among the hardest hit.

Read a blog about this here.

…

Walrus announced the launch of the MemWal NemoClaw and OpenClaw AI Agent Plugin, an agentic orchestration framework that allows AI to store durable memory on Walrus. The MemWal durable memory framework solves one of the most pressing issues for AI development, allowing agents to tap into instantly available and verifiable datasets for long-term task memory. 

…

WD has entered into exchange agreements with some institutional investors under which it will receive an aggregate of 1,865,801 shares of WD common stock in exchange for an aggregate of 653,203 shares of common stock of Sandisk Corporation. Upon the settlement of the exchanges, WD will continue to own 1,038,681 shares of Sandisk common stock, which it intends to dispose of in one or more subsequent exchanges for WD debt held by WD creditors and/or in exchange for outstanding shares of WD common stock and/or through distributions to WD stockholders as dividends.

…

ReRAM developer Weebit Nano said two product customers have successfully taped-out (released to manufacturing) chip designs intended for eventual mass production which integrate its ReRAM module. One customer has a prototype already manufactured and functional. Tape-out by product customers is an important milestone on the path to mass production and marks the achievement of one of the three 2026 targets set at Weebit’s 2025 Annual General Meeting.

Overlord Labs has integrated Weebit Nano’s ReRAM technology into the design of its next-generation smart battery management system, enabling significant gains in efficiency and reliability. The chip design was recently taped-out at DB HiTek, which once manufactured and shipped, will deliver advantages in power consumption, cost, and overall performance for high-volume applications. 

…

Xinnor announced the successful deployment of its xiRAID software at the Scientific Computing and Imaging (SCI) Institute at the University of Utah, replacing an aging single-drive PostgreSQL backend that had become a bottleneck for one of the world's largest research data environments, cataloging over 3 billion files across several petabytes of storage. After evaluating free alternatives including mdadm and ZFS, SCI selected xiRAID for a new server built around ten 3.84 TB enterprise NVMe drives in a RAID10 configuration. 

Ten raw NVMe drives running in parallel established the hardware ceiling at 8.03 million read IOPS and 125 GB/s. xiRAID RAID10 on a raw block device matched and slightly exceeded that ceiling, reaching 8.24 million read IOPS and 129 GB/s. Adding the ext4 filesystem on top, which is the actual production configuration, brought read performance to 7.71 million IOPS at 120 GB/s, representing only about 6% overhead versus the raw RAID device.

The old server, a single NVMe drive with ext4, managed 200K read IOPS and 3.1 GB/s. The new production system therefore delivers roughly 39 times more read IOPS and 5 times more write IOPS at peak, while keeping latency sub-millisecond under the same high-concurrency workloads that previously caused the old server's response times to collapse to 24 milliseconds.

The full case study is available at: https://xinnor.io/case-studies/sci-institute-university-of-utah/. 

…

Distributed AI database supplier Yugabyte launched Meko, an agent-native data infrastructure designed specifically for multi-agent AI systems to give agents the persistent, shared memory and knowledge they need to compound their learning over time. Meko introduces a new storage paradigm that gives AI developers a shared layer for memory, knowledge, conversation history, and observability, replacing the brittle stack of relational databases, vector stores, document stores, caches, and object storage, which IT teams are forced to stitch together.

Meko exposes agent-native actions, such as “add knowledge”, that directly represent the AI data constructs used by AI agents. Developers can now build and interact with these abstracted functions through standard interfaces (MCP) while Meko automatically manages how data is stored, indexed, and optimized across underlying storage systems. Meko is built on YugabyteDB, a horizontally scalable, PostgreSQL-compatible distributed database that natively supports SQL, NoSQL, vector, time-series, and graph queries. This means a single query can span multiple data models without stitching, unlocking higher performance and lower costs.

Developers can get started by requesting access at www.mekodata.ai or find out more by joining the YugabyteDB community on Discord at mekodata.ai/discord.