VAST Data and lakehouse analytics vendor Cloudera forged a strategic alliance to eliminate GPU data bottlenecks for AI workloads via an integrated end-to-end AI factory data pipeline.
Cloudera’s containerized suite covering data engineering, streaming, analytics, machine learning and AI services will integrate with VAST’s AI Operating System (AI OS) and its full data services stack including DataSpace, DataBase, DataStore, DataEngine, AgentEngine and InsightEngine, all built atop the firm’s DASE (Distributed and Shared Everything) storage architecture. Built on Nvidia’s AI Data Platform reference blueprint, AI OS supports on-premises, neocloud, GCP and Azure deployments. Two new modules, PolicyEngine and Tuning Engine, were rolled into AI OS this February and are set for full release before year-end.
VAST Data Co-Founder Jeff Denworth commented: “Nearly all enterprises already own ample data to power AI initiatives. The core hurdle lies in unlocking data value to establish uninterrupted pipelines for AI inference, model fine-tuning and data analytics to develop next-gen intelligent applications. By joining forces, Cloudera and VAST enable clients to construct unified AI factories that unify data, intelligence and infrastructure into one cohesive cross-hybrid AI operations platform.”
The two firms explain VAST’s AI OS converts dormant enterprise data into actionable AI-ready datasets, with Cloudera layering on robust data engineering, analytics, governance and AI tooling. Their joint solution delivers the following capabilities:
- An integrated AI factory framework spanning raw data intake all the way to model rollout
- Uniform operational workflows across on-prem data centers, private cloud and major public cloud platforms
- Resolution of GPU underutilization via ultra-high-throughput, low-latency data pipelines
- Markedly boosted compute efficiency by maintaining steady GPU utilization rates
- Scalable high-performance storage supporting structured, unstructured and multimodal data sets
- Secure private AI deployments equipped with enterprise-grade governance and compliance controls
Enterprises leveraging this stack can deploy production-ready AI systems that continuously convert internal data into usable insights for model training, inference, analytics and agentic use cases, covering standard, private and sovereign AI scenarios. Teams can speed up Apache Spark tasks using Nvidia cuDF; Cloudera Data Engineering workloads gain extra acceleration as Spark jobs tap into VAST’s GPU-boosted data services, further optimizing AI task execution.
Cloudera Chief Strategy Officer, Chief Business Officer and Applied AI General Manager Abhas Ricky stated: “Organizations pour billions of capital into GPU hardware, yet most fail to hit full hardware utilization due to data flow bottlenecks. Our collaboration with VAST resolves GPU starvation and lets customers build genuine AI factories with frictionless data flow from ingestion through to final insights.”
The combined Cloudera-VAST AI factory solution is now purchasable via both vendors’ enterprise sales channels and global partner networks. Additional details can be found at the linked resource page.
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
