The Ohio Supercomputer Center (OSC) has recently unveiled a new high-performance computing (HPC) cluster named Cardinal, which is poised to significantly enhance AI processing power for artificial intelligence (AI) research and education. Developed in partnership with industry giants Intel, Dell Technologies, and Nvidia, Cardinal represents a major step forward in the computational resources available to researchers and educators in Ohio.
Cardinal’s introduction is a response to the growing need for advanced computing power in AI, which spans a variety of fields such as scientific research, engineering, and biomedical studies. The new system is set to more than double the processing power and capacity of OSC’s previous system, the Owens Cluster, which has been operational since 2016.
- 756 Max Series CPU 9470 processors, which will provide 39,312 total CPU cores.
- 128 gigabytes (GB) HBM2e and 512 GB of DDR5 memory per node.
New HPC Cluster
The heart of Cardinal boasts 756 Intel Xeon CPU Max Series 9470 processors, which together provide a staggering 39,312 CPU cores. This formidable processing capability is further enhanced by each node’s 128 GB of cutting-edge HBM2E memory and 512 GB of DDR5 memory, ensuring that even the most demanding HPC and AI tasks are handled with efficiency.
The backbone of the Cardinal cluster is formed by Dell PowerEdge servers, which are known for their high performance and reliability. The system’s architecture has been carefully optimized to manage workloads effectively, which is essential for the successful integration of AI into research and industry.
Cardinal’s software environment is designed with user-friendliness in mind, supporting traditional x86 programming models to facilitate easy adoption. The cluster includes thirty-two nodes that are equipped with Nvidia H100 Tensor Core GPUs, each with 1 TB of memory and 400 Gbps NVIDIA Quantum-2 InfiniBand networking, enabling exceptionally fast data transfer speeds. Additionally, sixteen nodes are specifically designed for large symmetric multiprocessing (SMP) jobs, featuring 104 cores and 2 TB of DDR5 memory each.
The strategic choice of Intel’s Xeon CPU Max Series processors ensures that Cardinal is compatible with popular AI frameworks and libraries, allowing users to leverage the latest technology to advance their AI research without significant alterations to their existing workflows.
With the launch of the Cardinal HPC cluster, demonstrate their commitment to advancing the frontiers of research and education. Cardinal’s state-of-the-art hardware and software capabilities position it to support a broader range of use cases and foster innovation across various industries by enhancing AI capabilities. Here are some other articles you may find of interest on the subject of supercomputers.
- Fugaku Supercomputer delivered to Riken scientific research
- Hewlett Packard build new supercomputer factory
- NVIDIA Eos DGX AI supercomputer designed for AI applications
- BullSequana XH3000 supercomputer
- 65 ExaFLOP AI Supercomputer being built by AWS and NVIDIA
- University of Cambridge Dawn Phase 1 Supercomputer powers-up
- Meta AI supercomputer created using NVIDIA DGX A100
Latest Geeky Gadgets Deals
Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.