Giga Computing, an industry innovator and leader in enterprise hardware and advanced cooling solutions, today announced four new GIGABYTE servers built on the NVIDIA HGX B200 platform. This expansion of the GIGABYTE GPU server portfolio brings greater thermal design flexibility and support for the latest processors, including the new AI-optimized Intel Xeon 6 CPUs, giving customers more options as they tailor their systems for workloads and efficiency.
NVIDIA HGX B200 propels the data center into a new era of accelerating computing and generative AI. Built on NVIDIA Blackwell GPUs, the HGX B200 platform features 15X faster real-time inference on trillion-parameter models.
Liquid-cooled GIGABYTE G4L3 Series Servers
- G4L3-SD1-LAX5: supporting Intel Xeon CPUs
- G4L3-ZD1-LAX5: supporting AMD EPYC CPUs
The G4L3 Series is designed for compute density with dedicated CPU & GPU chambers. Engineered with advanced direct liquid cooling (DLC), the 4U liquid-cooled GIGABYTE servers offer incredible thermal efficiency and are ideal for high-performance AI training and inference at scale. Supporting the NVIDIA HGX B200 platform, maximizes GPU utilization while reducing power and cooling costs to make a future-ready platform.
Optimized Airflow in GIGABYTE G894 Series Servers
- G894-AD1-AAX5: supporting Intel Xeon 6900-Series processors
- G894-SD1-AAX5: supporting Intel Xeon 6700 & 6500-Series processors
The G894 Series is built to be readily adaptable to x86 platforms while supporting baseboard GPUs such as the NVIDIA HGX B200 platform. With the launch of new Intel Xeon CPUs that are optimized for AI host duty, the G894 series has expanded with two tailored server models. The G894-AD1 fully supports Intel Xeon 6900 series processors, including the Intel Xeon 6962P, while the G894-SD1 supports Intel Xeon 6700 & 6500 series processors, including the Intel Xeon 6776P.
New Servers Supported in GIGABYTE GIGAPOD
GIGAPOD is a rack-scale AI supercomputing solution purpose-built to tackle the immense computational demands of training generative AI and large language models with billions of parameters. By seamlessly integrating high-performance GPUs, high-speed networking, and scalable architecture, GIGAPOD delivers exceptional throughput and flexibility. Now featuring new GIGABYTE NVIDIA HGX B200 platforms, GIGAPOD introduces a refined design that enhances performance, improves energy efficiency, and provides customers with advanced cooling options to meet evolving AI workloads.
Source: GIGABYTE
