**Anomalo **introduced AIDA, its AI Data Analyst built directly on top of its data quality platform. With AIDA, every data user from analyst to executives can interact with their data using natural language, surfacing insights instantly and knowing that every answer comes from monitored high-quality datasets. AIDA was built in conjunction with Google Cloud, Databricks and Snowflake.

…

Cloud storage provider BackBlaze released a Network Stats Report, revealing how AI workloads are reshaping global data flows to the extent that the company is building a new metric. Backblaze’s Q3 2025 dataset highlights that traffic associated with “neoclouds” (cloud providers offering compute, GPU or other AI-related services) already represents nearly one-quarter of total ingress and egress across Backblaze’s global network. Historically, CDN traffic dominated as Backblaze’s B2 Object Storage offering grew by hosting enterprise backup and storage. Now, AI-related workloads represent a new class of traffic reshaping the company’s network profile.

Unlike traditional workloads, AI workloads move in short, intense bursts tied to dataset replication, model training, and inference pipelines. This activity is concentrated in a smaller number of IP addresses—fewer talkers, bigger flows—reflecting the high-density, high-throughput nature of AI compute clusters. Backblaze introduced a new metric: the “magnitude” of traffic, which reveals how much traffic each unique endpoint carries as a proxy for network intensity. The report shows that AI-related traffic flows are an order of magnitude more intense than traditional patterns, offering an early signal of how the infrastructure beneath the AI economy is taking shape.

The Q3 2025 Network Stats Report: The Magnitude of AI Workflows is available today on the Backblaze blog.

…

Data streamer Confluent announced Confluent Intelligence, the first unified platform designed to build and power context-rich, real-time AI, during its annual Current conference in New Orleans. Built on Confluent Cloud, the platform combines data streaming, processing, and AI reasoning in one place, enabling developers to build agentic AI systems that act on real-time data. Key capabilities include:

• Real-time context engine – Streams structured, reliable context to AI agents or applications via Kafka, Flink, or MCP, accelerating real-time AI development.
• Streaming agents – Lets teams build and run event-driven AI agents directly on Flink for intelligent, context-aware automation. Agents are natively integrated with Anthropic’s Claude.
• Built-in machine learning functions – Adds anomaly detection, forecasting, and model inference directly into Flink SQL, empowering faster, smarter insights.

…

Databricks has announced three major updates to its Agent Bricks platform:

• MLflow for Agent Quality (GA): Evaluate every agent interaction with custom scoring and human feedback tools.
• MCP Catalog in Marketplace (Preview): Secure access to trusted models and data from partners like Glean and You.com.
• ai_parse_document (Preview): Instantly extract insights from PDFs, tables, and other unstructured docs.

These tools aim to help enterprises build agents they can trust, with accuracy, governance and flexibility built-in.

…

Cloud resource monitor Datadog launched Storage Management, which helps prevent unexpected cloud object storage spend. It’s GA for Amazon S3, with previews for Google Cloud Storage and Azure Blob Storage. Uncovering specific savings opportunities—such as transitioning cold data from expensive storage classes, deleting duplicate objects or stopping the accumulation of non-current object versions—is a time-consuming, manual and reactive effort. Datadog Storage Management provides detailed insights into cloud object storage at the bucket and prefix levels across billions of objects. Learn more here.

…

Cosm is using** Dell Technologies **to redefine immersive entertainment with its Shared Reality, concept; think mini-Spheres minus the outside LEDs and with food and drink areas. Dell says we should imagine stepping into a massive LED dome where live sportslike College Football, the NBA, MLB and blockbuster films like The Matrix come alive in stunning 12K+ resolution with sound that surrounds you and hundreds of people sharing the same breathtaking moment. With the ability to stream up to 300TB of data seamlessly, PowerScale serves as the backbone for Cosm’s Shared Reality venues, delivering 12K+ LED visuals and wrap-around sound that bring live sports, blockbuster films, and educational journeys to life. PowerScale’s performance—offering 2x the efficiency per watt of comparable solutions—ensures Cosm can deliver real-time content integration, audience-adaptive visuals, and flawless streaming across its venues.

As Cosm scales to 100 global locations, each powered by its CX System LED dome technology – an end-to-end solution comprising an LED dome display, flexible and powerful immersive software, and wrap-around live and original content, Dell’s storage solutions are enabling rapid deployment and consistent, high-quality experiences worldwide. You can read more about the experience in a just-published Dell blog here and a YouTube video here.

…

Dell says it can offload AI processing KV Cache to its PowerScale and ObjectScale storage validated software stack, featuring the vLLM inference engine, LMCache with integrated Dell connector, and the NIXL data transfer library from NVIDIA Dynamo-extended by Dell with an S3-over-RDMA plugin for ObjectScale. It benchmarked this to measure Time to First Token (TTFT) performance with a 100% KV Cache hit rate. This isolates the efficiency of retrieving a fully prepopulated KV Cache from external storage. The tests were run on 4x NVDIA H100 GPUs across Dell storage backends. It compared its storage offload to a baseline where the KV Cache is computed from scratch on the GPU, using the LLaMA-3.3-70B Instruct, with Tensor Parallelism=4.

Dell PowerScale and ObjectScale delivered a 1-second TTFT at a full context window of 131K tokens; a 19x improvement over the standard vLLM configuration, which took over 17 seconds at the same context size. At a small context size of 1K tokens, the Dell engines retrieved the KV Cache from storage faster than could be computed on the GPU. More info in a blog.

…

Data mover Fivetran says Hippocratic AI has deployed Fivetran to generate accurate, real-time insights at scale without compromising sensitive healthcare systems, enabling clinicians to create reliable and secure healthcare agents. Hippocratic AI recently partnered with the UK’s Guy’s and St Thomas’ NHS Foundation Trust through the PATH initiative.

…

IBM Db2 Version 12.1.3 has added vector support – yes, old ponies can learn new tricks. It notes “There are databases in the marketplace that are dedicated to vector storage and efficient similarity search. However, these solutions have limitations. They rely on external data repositories. They have little or no ability to enhance queries with other filtering capabilities using the metadata associated with the vector embeddings. They have limited capability in other data management areas.

“Many Db2 customers already have the original input from which vectors are created right in their Db2 databases, so their vector storage solution is greatly simplified. The query capabilities of Db2 are unmatched, and the addition of vector similarity search enhances these capabilities. Db2 has developed industry-leading functionality in areas like scalability, security, auditability, performance, and monitoring over many years. Most vector-specific solutions just can’t offer that.” Read more here.

…

IBM has joined the Open Search Foundation as a Premier Member, and will announce and detail a new open source project featuring OpenSearch on the upcoming OpenRAG Summit livestream on Thursday, November 13. IBM aims to deepen integration between OpenSearch and its open source ecosystem, improving vector search performance, multimodal document ingestion and developer experience for AI agents. It plans to contribute enterprise-grade enhancements to OpenSearch’s observability and security solutions as well as high-availability patterns tested through IBM Cloud deployments.

DataStax, an IBM company, chose OpenSearch when it needed to build high-performance, production-ready vector search that could scale to billions of vectors without sacrificing recall or breaking budgets. By integrating Java-based vector search library JVector with OpenSearch through a custom plugin for dense vector retrieval, the achieved lower query latency and faster index builds, and reduced infrastructure overhead.

…

IBM announced IBM Fusion delivers one of the industry’s first implementations of the Nvidia AI Data Platform reference design that integrates enterprise storage with accelerated computing to power AI agents with near-real-time insights. One of the country’s leading academic medical centers, UT Southwestern, is using Nvidia’s AI Data Platform reference design, leveraging content-aware services to automatically prepare the medical center’s data for AI use. Clients can continuously process, index, and vectorize unstructured and semi-structured data to create AI-ready data for agents and derive insights to enhance the value of AI applications. IBM Fusion’s content-aware data services integrate with NeMo Retriever Microservices. Using RTX 6000 GPUs and Blackwell architecture, IBM Fusion is helping the medical center drive AI-driven medical research to support doctors, patients, and researchers.

…

InfluxData announced InfluxDB 3.6, available on both its Core and Enterprise products. Enhancements include:

• Version 1.4 of the Explorer UI, which includes **Ask AI, **a beta feature that lets users query data, documentation and system operations in natural language (no SQL required) and can perform operational tasks with single prompts. Version 1.4 also allows users to share dashboards between environments to make it easier to collaborate across teams.
• **Quick Start: **For development or local environments, users no longer need to configure instance or storage information before running the database. Quick Start is for developers who want to test InfluxDB without setup overhead, or want to run in a simple local environment.
• Major processing engine upgrades, including support for multifile Python plugins, direct uploads and updates, built-in observability, and tighter security controls. These improvements make code easier to maintain and reuse and allows users to iterate faster on complex data processing logic.

…

Salesforce-acquired data integrator **Informatica **announced new capabilities to connect enterprise data to AI agents:

• Six New CLAIRE Agents: autonomous agents to automate complex data tasks, including data quality rule creation, enterprise data exploration and pipeline building—reducing manual work from hours to minutes.
• Enhanced CLAIRE GPT: Upgraded with advanced reasoning and planning capabilities to help users automate and optimise data workflows through conversational interactions.
• AI Agent Engineering & Hub: No-code platform enables enterprises to build, test and deploy custom AI agents in minutes instead of weeks, with pre-built agents for Salesforce, Jira, Snowflake and more.
• Agentic Product Information Management: Simplifies product data management through conversational experiences, plus new Salesforce Agentforce integration that connects AI agents to unified master data.
• Enhanced Security & Governance: Introduces controlled unmasking for sensitive data access and multi-factor authentication for all native IDMC accounts to strengthen data protection.

…

In-memory, graph database supplier Memgraph has announced “two new tools specifically architected to open up the power of Retrieval-Augmented Generation based on graph technology (GraphRAG), to the entire database market, so democratizing GraphRAG access.” There is an immediately available AI Graph Toolkit—Unstructured2Graph and SQL2Graph — a set of open source libraries and utilities that automate the porting of both SQL and unstructured data into a knowledge graph in Memgraph, making it accessible to a chatbot running a GraphRAG pipeline. Later this month an MCP Client, within Memgraph Lab, that is fully compatible with the emerging standard for context engineering, will arrive. This is designed to make it easy to connect Memgraph’s graph capabilities to multiple back-end data sources through other MCP Servers, while supporting adoption of an industry standard to accelerate developer AI productivity.

The new Memgraph AI Toolkit is available for download through Github and can be used via Memgraph Cloud. For more technical information and background, read a blog.

…

Data protector NAKIVO achieved 20% booking growth and 15% collections growth in Q3 2025 Y/Y. EMEA accounted for 57% of booked revenue, Americas for 30%, and APJ for 13%. EMEA recorded the highest revenue growth of 29% Y/Y, with increased adoption among enterprise customers and MSPs. The largest deal came from Italy, valued at $74,884. The top two fastest-growing countries in terms of revenue were Greece and Belgium. Revenue doubled (100% growth) in several countries, including Guatemala, Kenya, Andorra, Greece, Bulgaria, Belgium, Ukraine, Montenegro, Denmark, Tunisia, New Zealand, India, Slovakia, Romania, Japan, Georgia, Indonesia, Maldives, and Poland. NAKIVO’s customer base expanded by 9% Y/Y, with customers in 188 countries, including, for the first time, Bahrain and Jersey. Notable new customers include Siena College, City of Orville, European Hospital Italy, and Tesla Engineering LTD. Enterprise deployments grew by 19% Y/Y, driven by adoption from larger organisations and MSPs. The company also welcomed 203 new partners globally in the quarter.

…

Analyst and research house Osmium Data Group released its first Trendscape on Data Protection, a 49-page insight report priced at €499.00 ($578.77). It covers strategic and innovative capabilities that organizations should expect from data protection solutions in 2025, and looks at offerings from Arcserve, Arctera, Asigra, Barracuda, Cobalt Iron, Cohesity, Commvault, Dell, Druva, HPE, HYCU, IBM, N-Able, Nakivo, Rubrik, Unitrends (Kaseya) and Veeam.

…

Pure Storage and Cisco announced a new converged infrastructure (CI) FlashStack Cisco Validated Design (CVD) as part of Cisco’s AI POD module forming a blueprint for an AI Factory. This FlashStack CVD adds in Nvidia GPUs to provide the AI angle and the whole thing is designed to help enterprise customers move from AI pilot projects to large-scale deployment via the AI factory idea, for use cases like RAG, agentic AI, semantic search, video analytics, and code generation.

This AI-focussed FlashStack CVD has Pure’s FlashBlade//S array; the “S” signifying performance-optimized, and its Portworx Kubernetes container orchestration software. Cisco provides its UCS C845a servers with Nvidia RTX PRO 6000 GPUs, and the system also includes Nvidia’s AI Enterprise software. It’s a validated design meaning that customers can buy it as a ready-made system instead of selecting, buying and integrating their own components. FlashStack CVD will be generally available in the first 2026 quarter through channel partners including AHEAD, ePlus, SHI International, and World Wide Technology. Customers can engage immediately through early access and briefings via Pure Storage, Cisco, or Nvidia field teams. Learn more here.

…

A Pure Storage [blog](https://blog.purestorage.com/solutions/modernize-pure-storage-cios-vmware-kubernetes/  -) says Portworx now provides a pathway to running VMs on Kubernetes. Its new Kube Datastore introduces a modern VM architecture optimized for cloud-native deployments, but familiar to existing skilled VMware administrators. By bringing familiar architecture and workflows to Kubernetes, we eliminate the steep learning curve that often slows modernization. Enterprises can run everything from lightweight edge VMs to heavy virtual desktop infrastructure (VDI) workloads on Kubernetes, without disruption or cyber risk. This helps CIOs confidently modernize by making Kubernetes a viable platform not only for new, cloud-native applications but also for traditional VM workloads.

…

The SNIA announced its fy 2026 board of directors exec committee:

• Chair: Dr. J Metz, Advanced Micro Devices, Inc. (AMD)
• Vice Chair: Richelle Ahlvers, Intel Corporation
• Secretary: Chris Lionetti, Hewlett Packard Enterprise (HPE)
• Treasurer and Advisor: Sue Amarin, Industry Consultant
• Member: Scott Shadley, Solidigm
• Advisor: Michael Meleedy, SNIA Chief Operations Officer

Learn more about the SNIA Board of Directors and Technical Council.

…

Amazon Prime Video has a production ecosystem for NBA on Prime with a 7-petabyte NVMe storage system from TrueNAS, enabling Prime Video to capture, store, and distribute live sports feeds across multiple continents with ultra-low latency. It supports simultaneous broadcasts from the new NBA on Prime studio in Los Angeles, mobile production units on both U.S. coasts, and control rooms in London that deliver NBA coverage to more than 220 countries in 12 languages.

…

**Tuxera **is demonstrating enterprise-grade file sharing for Linux servers at SC25. It’s providing a tech preview of Fusion NFS which offers RDMA support, scale-out architecture, and performance modernized for AI whilst providing greater flexibility than traditional NFSD implementations, and running in user mode for simplified deployment and management.

…

**Western Digital **wants AI data centres to buy more of its disk drives. Nigel Edwards, its VP WW Channel Sales & Marketing, says: “London’s £10 billion data centre boom marks a major leap in the UK’s digital infrastructure, and we will see hard disk drives (HDDs) at the heart of it. With AI and the exploding amount of data used and generated, hyperscale operators need high-capacity, reliable, and efficient storage at scale. HDDs are uniquely positioned to offer exactly that, while remaining powerful and performant at scale. …

“AI is transforming business across the country, but to fully tap into the potential of these new workloads, IT decision-makers need to look beyond processing power. Every model, agent, and tool depends on vast volumes of structured and unstructured data that need to be stored, processed, and analysed efficiently. In doing so, leaders often underestimate how much capacity their AI ambitions are consuming. For AI to succeed, a storage strategy must not be an afterthought.

“Today’s high-capacity HDDs enable data centres to store petabytes of information efficiently while keeping power and total cost of ownership in check. With AI driving London’s next wave of innovation, the city’s success will hinge on data infrastructure built to keep every insight – from real time to archive – online and within reach.”

WD also supplies its OpenFlex Data24 NVMe-oF Storage platform, with up to 1.474 TB of SSD storage capacity inside it.

…

Alison Cerra.

Veeam has appointed Alison Cerra as Chief Marketing Officer (CMO) and she will lead Veeam’s global marketing organisation, overseeing brand strategy, product marketing, demand generation, events, communications, and digital engagement. Cerra succeeds Rick Jackson, who is retiring after a 40-year career and three-and-a-half years as Veeam’s CMO. She comes from being Akamai’s CMO.

…

Cornerstone, a UK-based mobile and digital infrastructure services company, announced a strategic partnership with Stonesthro leveraging Zadara’s cloud technology to strengthen the UK’s digital resilience and sovereignty.