High Performance Computing
Making Data-Intensive Research Possible
The Cal Poly Pomona HPC Initiative supports research and instruction across colleges by providing access to high-performance computing resources. Whether you’re working with large datasets, running complex calculations, or getting started; we're here to help.
HPC Platforms
Hardware Specifications
Key Features
Polaris offers a seamless user experience combined with the necessary processing power to tackle complex problems efficiently. This platform provides the performance and storage required for demanding computational tasks.
Details
CPP’s newest high-performance computing cluster provides a modern, significantly more capable platform for advanced research, computational modeling, data analysis, and artificial intelligence. Polaris includes three compute nodes, each equipped with 160 CPU cores, 1.5 TB of RAM, and a GPU node featuring eight NVIDIA H200 GPUs with 3 TB of RAM. Polaris provides the performance, memory, and storage capacity required for demanding workloads including generative AI, large language models, deep learning, scientific simulations, and other data-intensive research.
Polaris provides increased computing resources within a smaller physical footprint than our Legacy HPC enabling researchers to tackle larger and more complex problems while using campus data center space and energy more efficiently. The system also provides a scalable foundation for expanding CPP’s research computing capabilities as faculty develop new interdisciplinary initiatives, grants, and partnerships.
Key Features
The Legacy system is built for efficiency and scale. It allows researchers to tackle larger and more complex problems using a traditional terminal interface.
Details
CPP's Legacy HPC consists of multiple dedicated processor nodes connected by a specialized high-speed network and job scheduling software, Its software management suite utilizes HP Enterprise's HPC Software Stack, including Slurm, an open-source job scheduler, Insight CMU for cluster management, and other HPE software for node deployment and configuration. The Anaconda package management system allows users to install and manage dedicated libraries and external software packages right from a terminal window.
The Slurm scheduler manages allocation, dispatching, and execution of jobs. Slurm is a well-documented resource manager allows tasks to be dispatched in real-time or batch mode.
Our Legacy HPC nodes are configured as partitions for dispatching jobs to appropriate nodes for computational tasks. The "General Compute Partition" is used for general-purpose jobs that benefit from running multiple computing tasks in parallel. The "GPU Partition" allows a task to access dedicated GPU processors where the task benefits from additional numerical processing capability.
The Legacy HPC cluster based on the HP's Proliant server platform includes two DL360 management nodes, 20 DL160 compute nodes, and four GPU nodes with eight Tesla P100 GPUs. The cluster offers 3.3TB of RAM connected through a dedicated internal 40GBit Infiniband switching fabric and 10GBit external ethernet. The overall system throughput is approximately 36.6 Tflp in double-precision mode or 149.6 Tflp in half-precision mode.