Celebrating 25 Years of Scoop
Special: Up To 25% Off Scoop Pro Learn More

Video | Agriculture | Confidence | Economy | Energy | Employment | Finance | Media | Property | RBNZ | Science | SOEs | Tax | Technology | Telecoms | Tourism | Transport | Search

 

VAST Data Achieves High-Performance Storage Solution Certification For NVIDIA Partner Network Cloud Partners

VAST Data, the AI data platform company, today announced that the VAST Data Platform has been certified as a high-performance storage solution for NVIDIA Partner Network cloud partners. This certification underscores VAST’s position as a leading data platform provider for AI cloud infrastructure, and further strengthens VAST’s collaboration with NVIDIA in building out next-generation AI factories.

VAST provides a unified set of storage and data services that empowers cloud service providers (CSPs) to offer a comprehensive catalogue of data-centric offerings that are deeply integrated with NVIDIA technologies. Uniquely designed to meet the stringent requirements of large-scale AI cloud infrastructure, the VAST Data Platform supports training and fine-tuning AI models of all sizes and modes, ranging from multimodal and small language models with less than ten billion parameters to the world’s largest models consisting of over one trillion parameters.

Advertisement - scroll to continue reading

Are you getting our free newsletter?

Subscribe to Scoop’s 'The Catch Up' our free weekly newsletter sent to your inbox every Monday with stories from across our network.

“This certification with NVIDIA builds on VAST’s already tremendous success with CSPs as the de facto AI data platform solution for large-scale, cloud infrastructure,” said John Mao, vice president, technology alliances at VAST Data. “With the VAST Data Platform independently validated and certified through the NVIDIA Partner Network, organisations can more confidently and securely deploy their AI models at unprecedented scale across thousands of GPUs.”

John Mao (Photo/Supplied)

With seamless scalability and industry-leading uptime, the VAST Data Platform enables service providers with capabilities that span the complete AI pipeline – from multi-protocol data ingest, accelerated data preprocessing for feature engineering, and high-performance storage for model training with fast checkpoints and restores to edge data management for model serving and end-to-end cataloguing to meet audit and compliance requirements.

The VAST Data Platform offers CSPs:

  • Service Provider-Grade Reliability: The VAST Data Platform helps observed production systems realise 99.9999 per cent availability.
  • Secure Multi-Tenancy: With a zero-trust framework, per tenant data encryption, flexible network segmentation, and robust audit capabilities, VAST enables CSPs to deliver secure cloud services at scale.
  • Fine-Grained Workload Isolation: With granular quality-of-service policies that prevent multi-tenant I/O contention, the VAST Data Platform helps CSPs ensure numerous customers have the performance and data access they need for AI workloads from a single cluster.
  • Infrastructure Cost Savings: Service providers and enterprises alike can consolidate storage silos with an affordable single all-flash solution and eliminate the overhead of data copies and sprawl.
  • Improved Operational Efficiency: VAST offers robust APIs and SDKs, and all Day 2 operations (such as upgrades) can be performed online, leading to a better customer experience with fewer admins required.

The VAST Data Platform serves as the comprehensive software infrastructure required to capture, catalogue, refine, enrich, and preserve data through real-time deep data analysis and learning. This empowers AI clouds to offer a diverse range of data services to their customers, further enhancing their AI capabilities.

© Scoop Media

Advertisement - scroll to continue reading
 
 
 
Business Headlines | Sci-Tech Headlines

 
 
 
 
 
 
 
 
 
 
 
 

Join Our Free Newsletter

Subscribe to Scoop’s 'The Catch Up' our free weekly newsletter sent to your inbox every Monday with stories from across our network.