Integrated Solution is Co-designed by SourceCode and Features Best-in-Class Technology from Gigabyte and VDURA for Simplified AI Factory Deployment
The Atlas AI Cluster is SourceCode’s latest innovation in collaboration with supercomputing leader Gigabyte and high-performance storage pioneer VDURA, designed to simplify and accelerate the development of AI factories. This integrated solution combines advanced engineering with best-in-class technologies to reduce cost and complexity while delivering leading performance, efficiency, and seamless scalability.
Building an AI factory requires more than selecting compute, storage, and networking. Atlas AI Cluster brings these technologies together in a co-designed, integrated architecture engineered around AI workload, performance, scalability, and deployment requirements. SourceCode helps customers simplify infrastructure decisions, accelerate deployment, and scale as AI requirements grow.
Atlas simplifies AI factory deployment with a co-designed architecture that integrates compute, storage, networking, and software, reducing integration complexity and accelerating time to production.
Highly configurable, Atlas is available in four configurations, incorporating high-bandwidth, low-latency NVIDIA Quantum-X800 800Gb/s InfiniBand networking, NVIDIA GB200 Grace Blackwell, and up to 13 PB of raw storage in the baseline configuration.
Leverages a full suite of management and AI software, including the option to use NVIDIA NIM microservices, the NVIDIA AI Enterprise software platform, the NVIDIA Omniverse platform, and Cloud, as well as optimized open-source foundational models.
SourceCode offers application assistance, optimization, knowledge transfer, cluster management software, and onsite integration to ensure seamless deployment.
The Atlas AI cluster is suitable for learning and inference across a wide range of AI use cases. It comes with the flexible configuration that is necessary to optimize it for specific applications such as:
Developed through a strategic partnership with Gigabyte and VDURA, Atlas represents the pinnacle of AI infrastructure. Â
The inclusion of the new VDURA Data Platform in Altas ensures AI-optimized metadata and input/output operations per second (IOPS), flexible deployment for maximum efficiency and simplicity, and enhanced data protection.
Atlas combines compute, high-performance storage, high-speed networking, AI software, and SourceCode engineering into an integrated architecture, reducing the complexity of designing, integrating, validating, and deploying AI infrastructure.
Atlas is highly configurable, allowing compute, storage, networking, and software to be aligned with specific training and inference requirements rather than relying on a one-size-fits-all architecture.
SourceCode provides engineering and application assistance, optimization, knowledge transfer, cluster management software, onsite integration, and deployment support.
