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.
Atlas is the quick way to build and scale an AI factory, helping you achieve time-to-value faster than traditional solutions.
Highly configurable, Atlas integrates up to 208 NVIDIA GH200 processors, 400G InfiniBand interconnect, and up to 13 PB of raw storage.
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.
