VAST Data integrates AI operating system with SK Telecom’s sovereign data center platform

VAST Data has announced a partnership with SK Telecom to integrate its AI Operating System into SKT’s new sovereign AI infrastructure, known as the Haein Cluster. The deployment utilizes the latest NVIDIA Blackwell graphics processing units (GPUs) and is targeted at accelerating national AI model development for South Korea. SK Telecom reports that the Haein Cluster is initially designed for GPU-as-a-Service (GPUaaS) and has been selected for Korea’s Ministry of Science and ICT’s “AI Computing Resource Utilization Enhancement (GPU Rental Support) Program.”

The Petasus AI Cloud, SKT’s virtualized infrastructure platform, incorporates the VAST AI Operating System with VAST’s disaggregated, shared-everything (DASE) architecture deployed on NVIDIA HGX servers manufactured by Supermicro. This configuration is designed to enable secure, multi-tenant workloads for both AI training and inference at scale. According to SK Telecom, this setup can provision GPU environments for users in under 10 minutes, with performance close to bare-metal, contrasting with the slower provisioning typical of traditional bare-metal data center infrastructure.

Key features of the integrated platform, as stated by VAST Data and SKT, include:

  • Isolation and control for sovereign-grade infrastructure, ensuring data and services remain within South Korea’s borders
  • Dynamic, virtualized provisioning of GPU resources with near bare-metal performance
  • Support for secure, multi-tenant AI workloads with privacy and performance guarantees
  • Unified pipelines for both AI model training and inference
  • Multi-protocol, client-agnostic access
  • Carrier-grade uptime and operational efficiency
  • Elastic scaling for national public and private sector AI workloads
  • Automated allocation and isolation of GPU and storage resources per tenant

“VAST Data’s unified architecture has been instrumental in helping us move from legacy bare-metal deployments to a fully virtualized, production-grade AI cloud,” said DK Lee, Vice President, and Head of the AI DC Lab at SK Telecom. “The VAST AI OS powers the performance, simplicity, and flexibility needed to support the next generation of sovereign AI workloads, and gives us the confidence to scale fast and securely. With VAST, we’re enabling a GPUaaS platform that meets the exacting needs of government, research, and enterprise AI customers in South Korea.”

The integrated solution, featuring Supermicro’s NVIDIA HGX-based servers, is positioned for deployment in data centers operated by telecom, research, and commercial providers that require high-throughput, sovereignty, and rapid deployment of AI resources within South Korea.

Source: VAST Data

Get Data Center Engineering News In Your Inbox:

Popular Posts:

Screenshot
Knowles launches safety-certified high-voltage ceramic capacitors for data center applications
2021_Celestial_logo-black
Celestial AI raises $255 million to deploy photonic interconnects for data center AI infrastructure
Yokogawa-T&M---SL2000-high-speed-data-acquisition-unit
Yokogawa releases SL2000 high-speed data acquisition unit for multi-channel measurement and analysis applications
Normal Computing Logo
Normal Computing announces thermodynamic data center chip CN101 for energy-efficient AI workloads
GmyybkpD3o7f45xka2pctu42nog
FS launches high-density 1U WDM platform for data centers, maximizing rack space and passive cooling

Share Your Data Center Engineering News

Do you have a new product announcement, webinar, whitepaper, or article topic? 

Get Data Center Engineering News In Your Inbox: