Acceldata, provider of the xLake data platform enabling autonomous enterprise analytics and AI agent operations across cloud, on-prem, hybrid and sovereign environments, expands its European business. It recruits Ali Afsar, ex-HSBC SVP of Data Strategy, and Colin Fernandes (formerly at Sumo Logic and VMware) as Director of EMEA Market Development & Global Sales Enablement. EMEA VP Justyn Goodenough continues hiring to scale the regional field team.
Cohesity’s CTO Anubhav Gupta leaves after nearly a decade to launch his own startup; the firm thanked him for his contributions.
Cyber-resilience vendor Commvault’s sponsored July 2026 IDC whitepaper Resilience Operations: The Discipline that Makes Readiness Provable reveals key findings around AI agent risks, framing AI agents as digital staff. 90% of respondents aim to upgrade identity management systems to mitigate AI agent risks, and 58.7% plan major identity framework overhauls. Historical research shows 69% of global firms faced identity breaches between 2022 and 2025. Additional survey data:
- ---98.4% require tighter IT-security team collaboration
- ---Only 26.7% deploy dynamic AI-compatible role-based access control (RBAC)
- ---57.7% lack fully defined Minimum Viable Business IT recovery standards
Commvault also rolled out its sponsored IDC Cyber Readiness Assessment, an interactive tool to evaluate AI-era organizational resilience. The whitepaper is available now, while the assessment tool will launch within months.
CTERA rebranded its Enterprise File Services Platform to the CTERA Intelligent Data Platform, winning two Q2 2026 PeerSpot Customer Choice Tech Leader Awards, ranking second in Cloud Storage and Cloud Backup. Unlike paid analyst rankings, these awards rely solely on one year of verified reviews from enterprise IT practitioners. The honors recognize CTERA’s strengths in simplifying distributed file workflows, boosting cyber resilience, and building secure, compliant foundations for AI and analytics.
Keith Townsend of The CTO Advisor built Layer2C, an independent AI infrastructure evaluation website built around the 4+1 Layer AI Infrastructure Model. The framework maps decision authority across the full AI stack (compute, networking, data, orchestration, runtime, reasoning, applications), helping IT leaders distinguish vendor-managed functions from internal organizational responsibilities for governance, costs, audit trails and accountability. The platform evaluates 25 vendors, scoring each capability via the Decision Authority Placement Model (DAPM) into Retained, Delegated or Ceded categories and supporting side-by-side vendor comparisons.
Citrini analyst Jukan reported Chinese DRAM maker CXMT evicted Huawei engineers from its cleanroom facility following Huawei’s strong pushback against CXMT memory price hikes.
Databricks CEO Ali Ghodsi explained the firm’s latest fundraising round stems from extreme global GPU shortages. The company hosts open-source models including Kimi for clients, yet Asia GPU capacity is nearly exhausted with rising demand across Japan, South Korea, the U.S. and India. Massive, costly GPU procurement requires substantial new capital to fulfill overwhelming customer orders.
Databricks and Microsoft extended their decade-long strategic partnership through the 2030s to help enterprises build context-rich scalable AI deployments. Three core expansion pillars:
- Deeper Microsoft ecosystem integrations for Databricks Genie and Unity AI Gateway across Microsoft 365, Teams, Copilot, Power BI and Microsoft Foundry to embed trusted corporate data into daily AI workflows.
- Databricks will run its core operations and analytics workloads on Azure Databricks to validate the platform’s enterprise reliability.
- Expanded adoption of Microsoft Azure Cobalt hardware to accelerate AI agent and data-heavy workload performance and efficiency.
The joint solution delivers customized enterprise AI by linking models and agents to verified business data while retaining unified governance, security and cost oversight.
DCIG released a new competitive intelligence report comparing AI storage performance between HPE Alletra Storage MP X10000 and Dell PowerScale.
Data virtualization provider Denodo launched Platform v95.5. Its upgraded enterprise knowledge graph within the Denodo Data Marketplace enhances data context visibility to strengthen AI trustworthiness alongside standardized metric views for its semantic layer. The Denodo Assistant features improved reasoning logic, plus broader cross-stack data and AI connectivity to unify disjointed data sources and elevate business decision quality.
UK analytics consultancy GlobalData adopted the Model Context Protocol (MCP) across all its data and AI products. Modern business users demand embedded intelligence within existing AI tools, CRMs and copilots instead of separate standalone platforms, while data teams require structured, model-ready analytical outputs. MCP enables secure, compliant API integrations connecting enterprise AI systems to GlobalData’s proprietary datasets, shifting the brand from a standalone portal provider to embedded decision intelligence supplier.
Independent SaaS backup specialist Keepit joined the Pax8 global cloud and AI marketplace targeting SMBs. Managed service providers and channel partners via Pax8 can now deliver Keepit’s dedicated cloud backup and recovery platform for business-critical SaaS applications.
A federal jury in Waco, Texas ordered Kioxia to pay satellite firm Viasat $229 million for flash memory patent infringement. Viasat created low-power, long-lifespan flash error correction technology for satellite systems and claims Kioxia’s storage drives replicate its patented logic. Viasat also filed an ongoing parallel patent lawsuit against Western Digital (Sandisk).
Software-defined block storage developer Lightbits Labs rolled out Ubuntu OS compatibility, letting clients build high-performance disaggregated storage clusters without sacrificing speed, scalability or ease of operation; free product trials are available.
Composable infrastructure vendor Liqid formed a strategic partnership with AMD to develop next-gen AI infrastructure. The integrated solution combines Liqid UltraStack 30 GPU pooling technology with AMD PCIe Instinct GPUs, enabling servers to dynamically pool up to 30 GPUs delivering 60 FP8 PFLOPS and over 4TB aggregated HBM3E memory to run large-scale AI models. This pooled accelerator fabric delivers three key benefits:
- ---Up to 3.7× higher tokens per second inference throughput vs. traditional servers
- ---Up to 2.1× more AI tokens generated per dollar invested
- ---Up to 1.8× improved tokens per watt energy efficiency
Liqid’s UltraStack 30 platform supports future CXL memory pooling for cross-server shared memory expansion.
Micron signed long-term Strategic Customer Agreements (LTAs/SCAs) with seven top automotive Tier 1 suppliers: Qualcomm, Visteon, HARMAN, JOYNEXT, DENSO, Astemo and Hyundai Mobis. This follows prior supply pacts with GM and Ford, extending guaranteed advanced memory and storage access deeper into software-defined automotive supply chains for infotainment, ADAS and connected vehicle systems. The multi-year contracts lock in stable supply, transparent pricing and joint product roadmapping aligned with automotive hardware lifecycles.
Award-winning timber structural engineering firm StructureCraft migrated all file shares and virtualized infrastructure to NetApp to support global remote teams and AI-powered design software like Rhino 3D. Its distributed workflows generate massive file-heavy datasets that outgrew legacy storage. The completed deployment uses hourly NetApp Snapshots for fast design file recovery, centralizing operations to cut cross-continental data transfer costs. Long-term plans include building internal StructureCraft OS, a custom framework for staff to build proprietary AI tools atop NetApp storage.
ShelterZoom subsidiary Mithra Technologies debuted Mithra AI, cryptographic enterprise AI trust infrastructure that validates all AI outputs against trusted internal data as a model-agnostic Single Source of Truth (SSOT). Deployable via APIs, SDKs, OEM partnerships or full on-prem systems, it operates alongside major LLMs including OpenAI, Claude, Gemini, Copilot and Grok rather than replacing them. Pre-packaged as an AI-native content vault for corporate intellectual property, it caters to heavily regulated verticals such as healthcare, government, finance and legal sectors. Conscience IQ founder Kirby de Lanerolle confirmed active deployments across healthcare and defense clients.
Multi-tier real-time storage array developer StorONE partnered with Esprinet Group distributor V-Valley to expand its enterprise storage footprint in Italy. V-Valley will add StorONE to its data management portfolio and supply channel partners with training, pre-sales support, business development and lead generation resources. StorONE EVP of International Sales Zion Gabai noted V-Valley’s robust distribution network and technical expertise will help resellers deliver differentiated storage solutions to solve core customer pain points.
Market research firm TrendForce stated leading chipmakers and cloud service providers rolled out hardware and software innovations to ease 2026 H1 memory bottlenecks driven by surging KV Cache demand and tight memory supply. Hardware solutions include KV Cache servers and CXL switches from Penguin Solutions, Marvell and Meta to expand usable memory capacity, while Nvidia and Google built KVTC and TurboQuant compression tools to cut KV Cache memory consumption. TrendForce’s full report Overcoming the Memory Bottleneck: CXL Expansion and KV Cache Compression Innovations offers comprehensive technical and market analysis.
Archive technology provider Versity’s ScoutAM platform secured German Climate Computing Center (DKRZ) as a new customer. DKRZ runs Germany’s national climate supercomputing facility supporting over 1,000 climate and earth science researchers, collaborating with NEC to deploy a new hierarchical storage and tape archive system housing one of the world’s largest climate simulation datasets. Its Levante supercomputer features a 130 PB Lustre high-performance file system for active model processing and research data.
ScoutAM transfers data between DKRZ’s HPC clusters and archives at peak 30 GB/s throughput (over 2 PB daily). Its 8.2 PB combined SSD/HDD cache absorbs heavy data ingestion loads and keeps frequently accessed archive data instantly retrievable without tape loading, enabling the archive to act as an active storage tier. Data is offloaded from Lustre to free premium high-performance storage, while cached datasets avoid slow physical tape retrieval.
Beijing Qianxing Jietong Technology Co., Ltd.
Sandy Yang/Global Strategy Director
WhatsApp / WeChat: +86 13426366826
Email: yangyd@qianxingdata.com
Website: www.qianxingdata.com/www.storagesserver.com
Business Focus:
ICT Product Distribution/System Integration & Services/Infrastructure Solutions
With 20+ years of IT distribution experience, we partner with leading global brands to deliver reliable products and professional services.
“Using Technology to Build an Intelligent World”Your Trusted ICT Product Service Provider!
Your message must be between 20-3,000 characters!
English
