MSI Launches Advanced NVIDIA GPU Server Systems for AI and HPC
On August 31, MSI unveiled two powerful NVIDIA GPU server systems designed to elevate AI and simulation capabilities for enterprises. The new models, the 4U CG480-S5063 and the 2U CG290-S3063, are built on the innovative NVIDIA MGX modular architecture, aimed at enhancing performance without necessitating extensive data center overhauls.
Elevated Performance with NVIDIA RTX PRO 6000 Blackwell GPUs
Both server systems are equipped with the NVIDIA RTX PRO 6000 Blackwell server GPU version, offering state-of-the-art capabilities that enable businesses to improve their artificial intelligence (AI) and high-performance computing (HPC) tasks. This powerful combination is tailored to meet the demands of modern enterprises seeking efficiency and scalability.
Features of the 4U NVIDIA MGX Server CG480-S5063
The CG480-S5063 model is designed specifically for computing-intensive AI and HPC deployments. Here are some of its key features:
- Optimized Configuration: Supports a versatile CPU:GPU:NIC topology (2-8-5-200) that allows for expansive custom setups.
- Robust Processing Power: Comes equipped with dual Intel Xeon 6500/6700 processors, each with a TDP of up to 350W, to handle high workloads smoothly.
- Memory Capacity: Offers support for up to 8 TB of DDR5 memory via 32 DIMM slots, supporting speeds of 6400MT/s (1DPC) and 5200MT/s (2DPC).
- Storage Options: Features 20 front-facing hot-swap PCIe Gen 5.0 NVMe slots along with two M.2 2280/22110 PCIe 5.0 x2 NVMe slots, providing abundant storage solutions for data-intensive tasks.
- Expansion Flexibility: Includes up to 13 PCIe 5.0 x16 expansion slots, allowing users to incorporate up to eight 600W dual-width GPUs, including the RTX PRO 6000 Blackwell.
- Dedicated Management: Equipped with dual 1000Base-T management ports and the ASPEED AST2600 chip to facilitate easy monitoring and management, along with support for IPMI 2.0 and DMTF Redfish.
- Optimized for Large Scale: Ideal for extensive training, inference, and simulation tasks, making it a powerful asset for enterprises focused on AI advancements.
Features of the 2U NVIDIA MGX Server CG290-S3063
The CG290-S3063 model emphasizes efficiency and compactness, making it suitable for smaller spaces while still delivering high performance. Key features include:
- Inference and Model Training: Tailored for inference-oriented tasks, small model training, and fine-tuning, making it ideal for organizations with specific AI needs.
- Scalable GPU Configuration: Supports a flexible GPU topology of 1-4-3, allowing for diverse configurations.
- Processor Power: Houses a single Intel Xeon 6500P / 6700P processor, also with a TDP of up to 350W, balancing performance and energy efficiency.
- Expandable Memory and Storage: Provides 16 DDR5 DIMM slots with access to 136 PCIe channels, enabling flexible scaling for memory needs.
- Compact Design: Comes equipped with four rear PCIe 5.0 U.2 NVMe and two M.2 hard disk slots, facilitating effective storage management in a limited footprint.
- Optimized for Energy Efficiency: This compact 2U design strikes a balance between computing power and energy efficiency, making it a suitable option for expanding AI workloads in businesses where space is a constraint.
Conclusion
MSI’s introduction of the CG480-S5063 and CG290-S3063 servers marks a significant advancement in the realm of AI technology and HPC solutions. By harnessing the power of NVIDIA’s innovative GPU offerings, these new models not only optimize performance but also pave the way for enterprises to scale their AI capabilities smoothly and efficiently. As AI continues to reshape industries, such high-performance server systems will be pivotal in driving innovation and operational success.
By focusing on features and benefits, this SEO-optimized article is designed to inform and attract potential customers and tech enthusiasts interested in the latest developments in NVIDIA GPU server technology.
