cd /news/ai-infrastructure/ddn-gets-infinia-ready-for-productio… · home topics ai-infrastructure article
[ARTICLE · art-51201] src=blocksandfiles.com ↗ pub= topic=ai-infrastructure verified=true sentiment=↑ positive

DDN gets Infinia ready for production AI inferencing

DDN released Infinia v2.4, an object storage software update designed to optimize AI inference at scale, retrieval-augmented generation, and multi-protocol workloads. The update adds advanced multi-tenancy, identity management, quota enforcement, and initial POSIX support to help enterprises reduce inference costs and improve GPU utilization. DDN CEO Alex Bouzari emphasized that the economics of AI, measured by cost-per-token and inference efficiency, are now more critical than model performance.

read3 min views1 publishedJul 8, 2026
DDN gets Infinia ready for production AI inferencing
Image: Blocksandfiles (auto-discovered)

object

DDN has updated its Infinia object storage SW to v2.4, saying that this is not a minor, maintenance, release. It represents a strategic expansion into AI Inference at scale, RAG and intelligent data lakes, and multi-protocol workloads for research, life sciences, FSI (Fluid Structure Interaction) and manufacturing

It asserts that, as enterprises increasingly deploy agentic AI, retrieval-augmented generation (RAG), copilots, and autonomous AI systems, inference has emerged as the dominant operational cost in modern AI environments. It claims that Infinia's architecture is designed to optimize production inference economics through ultra-low-latency data access, high-performance object storage, and intelligent data services to keep GPUs processing data instead of waiting for storage.

The v2.4 Infinia update adds a set of features to help customers, such as Nvidia cloud partners, managed AI services providers, and enterprise AI users, use Infinia smoothly in their existing environments, supporting multiple teams, customers, business units, or sovereign AI workloads from a single platform.

DDN CEO and co-founder Alex Bouzari said: "The economics of AI are rapidly becoming more important than the models themselves. The industry has entered an era where success is measured by cost-per-token, inference efficiency, GPU utilization, and business outcomes—not simply the number of GPUs deployed. Organizations need infrastructure that transforms expensive AI investments into productive AI factories. Infinia 2.4 provides the governance, security, performance, and operational foundation required to maximize AI ROI while enabling the next generation of enterprise and sovereign AI.”

DDN is adding:

Advanced multi-tenancy

Identity integration and management

Quota enforcement and governance controls

Enhanced operational isolation for shared AI environments

Full compatibility with established S3 environments and SDKs

Initial POSIX support with POSIX client qualification on Red Hat Enterprise Linux and Ubuntu, defined throughput commitments, and documented deployment guidance and operational parameters.

Multi-tenancy being advanced is justified by tenants and subtenants being provisioned in ~10 seconds via REST API calls. Data objects are tagged with tenant IDs for isolation throughout the system and this supports hierarchical structures (tenants → subtenants → datasets) for flexible sharing among teams, customers, business units, or sovereign workloads on one cluster.

The relevance of POSIX is, according to a blog by Sanjay Jagad, VP of Product Management, DDN: “A genomics pipeline, a computational fluid dynamics simulation, a quantitative research workflow, a digital twin environment — these require standard POSIX file semantics. Until now, serving these workloads alongside AI training and inference meant separate infrastructure, separate operations, and data copies moving between them.”

“With native POSIX file access entering customer validation, Infinia extends its reach to these workloads without requiring separate infrastructure. File-based applications run on the same platform, against the same data, with the same performance and isolation guarantees as the AI-native object workloads Infinia already serves.”

Jagad says: ‘The reason Infinia can expand into inference, intelligent data lakes, and multi-protocol workloads simultaneously — without becoming a different product for each — is the architecture.

“A distributed, log-structured KV engine where every IO is independently optimized. Metadata in the same engine as data, guaranteeing real-time catalog freshness. Layout indirection that enables zero-downtime cluster scaling without moving a byte of data. Per-IO dynamic erasure coding that gives every write optimal protection without fixed stripe penalties. Hierarchical key-space multi-tenancy that provisions isolated tenants in seconds.”

He says: “Infinia can run training, inference, RAG retrieval, and POSIX file workloads on the same cluster, with hard isolation between tenants.”

Infinia’s roadmap looks like this:

Scale validation: 500–600 node clusters with published linear benchmarks

POSIX GA: Production-ready with GPUDirect Storage, POSIX ACLs, S3/POSIX interop on the same dataset

Enterprise security: RBAC with AD/LDAP, KMS/KMIP with per-tenant encryption keys, WORM/Object Lock

Data protection: Replication engine and data mover for business continuity

NVIDIA certification: DGX positioning and a formal certification path

Details of the full Nvidia integration roadmap are available to qualified customers and partners under NDA.

── more in #ai-infrastructure 4 stories · sorted by recency
── more on @ddn 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/ddn-gets-infinia-rea…] indexed:0 read:3min 2026-07-08 ·