The storage stories enterprise IT leaders actually read.
Volume 342, Issue 1 · 7 articles from 3 issues
StorageReview ranks six data center drives across mixed-use, read-intensive, high-capacity and EDSFF categories on lab data.
StorageReview publishes its enterprise SSD leaderboard for 2026, ranking six data center drives. Categories cover mixed-use, read-intensive, high-capacity and EDSFF form factor drives.
Rankings are based on the publication's current enterprise bench using FIO, GDSIO and DLIO, with drives tested on the older suite noted separately under Also Tested.
Every drive ranked has been through the StorageReview lab, and the publication states it does not rank announced products it has not tested. The leaderboard is aimed at buyers specifying data center flash.
5 Reasons NFS Makes the Best Default File Protocol for Backup
DCIG recommends organizations adopt NFS as their default file protocol for backup traffic due to its broad storage support, simplicity, and natural integration with Linux-based infrastructure.
While modern SMB 3.x offers competitive advantages through multichannel connections and RDMA capabilities, organizations should switch to SMB only when specific application requirements, Windows integration needs, or demonstrated performance benefits justify the change.
Protocol implementation and overall system architecture matter more than protocol selection alone, and organizations should prioritize meeting performance requirements over achieving theoretical maximum throughput.
Data sovereignty has moved from a compliance checkbox to a board-level resilience and portability question.
The piece argues data sovereignty is no longer just about where data is physically stored for regulatory purposes. Across multicloud, hybrid and AI service estates, enterprises need control over where data resides, who can access it, and how quickly workloads can be relocated.
With the global average breach cost cited at US$4.9 million, governance of data location and access has become a board-level concern.
Geopolitical tension, shifting regulation and architectural complexity make sovereignty a matter of operational flexibility and resilience rather than paperwork.
Governance fails in the gap between designed systems and what people actually do day to day.
Daragh O Brien argues data governance fails less because systems are badly designed than because of the gap between formal structures and actual human practice.
Where Stafford Beer's POSIWID explains failure after the fact, O Brien proposes POSIWPD - the practice of the system is what people do - since governance is continuously enacted by individuals interpreting their roles through professional traditions.
Drawing on Weick and Roberts' work on high-reliability organizations, he describes heedful interrelating and collective mind as what makes governance effective, citing studies finding it succeeds through coordination routines rather than artifacts.
He warns that automating routine work strips practitioners of the exposure needed for situational awareness during novel crises.
Hypervisor migration is forcing enterprises to evaluate storage and virtualization platforms as one decision.
Gartner's latest Magic Quadrant finds that hypervisor replatforming has opened a major displacement window in which storage and virtualization strategy must be reconsidered together.
By 2029, consumption-based storage services are projected to replace 50% of on-premises enterprise storage capex, while AI workloads drive demand for intelligent data platforms.
Rising hardware costs and supply constraints are expected to persist through 2028, making tiered storage architectures and automated data placement increasingly important for cost optimization across hybrid environments.
StorageNewsletter challenges the methodology and vendor inclusion criteria behind Gartner's 2026 storage quadrant.
StorageNewsletter published a critical analysis of Gartner's 2026 Magic Quadrant for Enterprise Storage Platforms, arguing the methodology lacks accuracy and consistency.
It notes that significant vendors including Vast Data, Lenovo, and Fujitsu are absent, while the category definition emphasizes interfaces over capacity.
The publication questions inclusion criteria such as the $325M revenue threshold and the minimum storage capacities, contending they fail to reflect real enterprise requirements or the diversity of available technology.
Deploying a DSPM tool is not a program; coverage scope, ownership, and continuous discovery are what matter.
A tool that produces a 10,000-finding inventory still leaves sensitive data undiscovered six months later, because the failure is program design rather than tooling.
Coverage scope must be defined explicitly before discovery runs, including test environments holding production copies and cloud resources provisioned outside scope.
Most environments need all three classification methods - content-based, context-based, and label inheritance - since each misses cases the others catch. Hybrid coverage combining continuous discovery for high-change environments with periodic scanning for stable ones works best.
Classification output must then feed access governance, DLP policy, retention, and incident response.