Leap Ahead of the Competition With GPU-Accelerated Computing
Moving to a GPU-accelerated strategy allows superior performance by all measures including faster compute time and reduced hardware requirements.
Get Ready for the Future with More Powerful, More Efficient Computing
Having deployed the world’s first HPC cluster powered by AMD and being named NVIDIA's HPC Preferred OEM Partner of the Year multiple times, the Penguin Solutions team is uniquely experienced with building both CPU and GPU-based systems as well as the storage subsystems required for AI/ML architectures and high-performance computing (HPC) and data analytics.
As AI continues to evolve and reshape industries, the role of expert partners in guiding organizations through implementing private AI infrastructures is increasingly vital. Our solutions shorten time to insight and discovery by removing the complexities involved in designing, deploying, and supporting customer’s AI infrastructure.
Our GPU-accelerated compute delivers best of breed solution that powers our technology practices. Our infrastructure offering includes both 19” EIA and 21” OPC servers enabling higher density, and alternative non-air cooling for more compute power per rack.
Additional reading on GPU-accelerators:
Computing Power/Speed
A single GPU can offer the acceleration performance of hundreds of CPUs for certain workloads.
Efficiency/Cost
A single GPU-accelerated server costs much less in upfront, capital expenses and, because less equipment is required, reduces carbon footprint and operational costs.
Flexibility
The inherently flexible nature of GPU programmability allows new algorithms to be developed and deployed quickly across a variety of industries.
Long-Term Benefits
Reliance on GPU-accelerated computing provides not only greater computing power over time, but greater margin of difference over competitors who do not migrate to GPU-accelerated computing.
Supported Applications
Amber, ANSYS Fluent, Gaussian, Gromacs, LS-DYNA, NAMD, OpenFOAM, Simulia Abaqus, VASP, WRF
Deep Learning Frameworks
Caffe2, Microsoft Cognitive Toolkit, MXNET, Pytorch, TensorFlow, Theano
19-inch EIA Servers
Altus XE2318GT
A step above, the Penguin Solutions® Altus® XE2318GT server is a GPU powerhouse, offering double the amount of GPUs from our previous generation. Leveraging the latest computing technology from AMD, paired with fast and efficient memory and just the right amount of storage, the Altus XE2318GT is the optimal 2U platform offering high-density GPU availability for large-scale enterprise and hyperscale deployments.
NVIDIA H100 NVL, L40, L40S
Relion XE2318GT
As more organizations look to grow their compute capabilities, they look to GPU accelerators to do so. Deploying GPU’s allows you to offload compute-intensive functions from CPUs, massively improving processing times. This is why workloads like artificial intelligence and deep learning, which need immense computing resources to process extreme amounts of data, benefit from GPU density. The Penguin Solutions® Relion® XE2318GT server packs 8 GPUs in a 2U rackmount form factor along with two 4th Gen Intel® Xeon® Scalable Processors. Choose the Relion® XE2318GT server to power data-intensive workloads that need the computing power of GPU accelerators.
NVIDIA H100 NVL, L40, L40S
Altus XE3314GTS
Graphics Processing Units (GPUs) have quickly established themselves as the computing tools of the future. As the growth rate of Central Processing Unit (CPU) speeds slows down, GPUs can offload compute-intensive functions from CPUs and accelerate your time-to-results. Penguin Solutions® Altus® XE3314GTS server is ideal for anyone who needs to maximize P2P GPU performance in a compact platform.
NVIDIA HGX H100 80GB HBM3 SXM5 x4, NVIDIA HGX H200 141GB HBM3e SXM5 x4
Relion XE3314GTS
Graphics Processing Units (GPUs) have quickly established themselves as the computing tools of the future. As the growth rate of Central Processing Unit (CPU) speeds slows down, GPUs can offload compute-intensive functions from CPUs and accelerate your time-to-results. Penguin Solutions® Relion® XE3314GTS server ideal for anyone that needs to maximize P2P GPU performance in a compact platform.
NVIDIA HGX H100 80GB HBM3 SXM5 x4, NVIDIA HGX H200 141GB HBM3e SXM5 x4
Altus XE4318GT
The Penguin Solutions® Altus® XE4318GT is a GPU powerhouse. Leveraging the latest compute technology from AMD, paired with fast and efficient memory, and flexibility in storage, the Altus XE4318GT is a 4U high-density GPU platform that can tackle any large-scale enterprise and hyperscale deployment.
NVIDIA H100 NVL, L40, L40S
Relion XE4318GT
If your organization is like many others, choosing to build its cluster around graphics processing units (GPUs), accelerate your time-to-results with the Penguin Solutions® Relion® XE4318GT server and its 8 GPU capacity. Choose the Relion® XE4318GT server for complex compute applications and mission-critical projects requiring the most powerful 4th Gen Intel® Xeon® Scalable Processors and the latest GPU-accelerator technology.
NVIDIA H100 NVL, L40, L40S
Altus XE5318GTS
Graphics Processing Units (GPUs) have quickly established themselves as the computing tools of the future. As the growth rate of Central Processing Unit (CPU) speeds slows down, GPUs can offload compute-intensive functions from CPUs and accelerate your time-to-results. The Penguin Solutions® Altus® XE5318GTS server is built on the latest technology offering the maximum amount of interconnected GPUs to process intense amounts of data with the least amount of latency.
NVIDIA HGX H100 80GB HBM3 SXM5 x8, NVIDIA HGX H200 141GB HBM3e SXM5 x8
Relion XE5318GTS
Graphics Processing Units (GPUs) have quickly established themselves as the computing tools of the future. As the growth rate of Central Processing Unit (CPU) speeds slows down, GPUs can offload compute-intensive functions from CPUs and accelerate your time-to-results. The Penguin Solutions® Relion® XE5318GTS server is built on the latest technology offering the maximum amount of interconnected GPUs to process intense amounts of data with the least amount of latency.
NVIDIA HGX H100 80GB HBM3 SXM5 x8, NVIDIA HGX H200 141GB HBM3e SXM5 x8
Altus XE5318GTO
Graphics Processing Units (GPUs) have quickly established themselves as the computing tools for the most demanding workloads. As the demands of advanced applications place on Central Processing Unit (CPU) increase, GPUs can offload compute-intensive functions from CPUs and accelerate your time-to-results. The Penguin Solutions® Altus® XE5318GTO server is built on the latest GPU technology to offer even greater levels of performance and speed to process intense amounts of data with the least amount of latency.
AMD Instinct MI300X 192GB HBM3 OAM x8
Relion XE5318GTO
Graphics Processing Units (GPUs) have quickly established themselves as the computing tools of the future. As the demands of advanced applications place on Central Processing Units (CPU) increase, GPUs can offload compute-intensive functions from CPUs and accelerate your time-to-results. The Penguin Solutions® Relion® XE5318GTO server is built on the latest GPU technology to offer even greater levels of performance and speed to process intense amounts of data with the least amount of latency.
AMD Instinct MI300X 192GB HBM3 OAM x8
Talk to the Experts at
Penguin Solutions
Our strong history and relationships with AMD and NVIDIA, combined with our extensive experience in AI and HPC infrastructure, allow us to provide GPU-accelerated cluster solutions tailored for every customer.
Reach out today and let's discuss your AI infrastructure requirements.