- Complete high-performing AI architectures based on Penguin Solutions OriginAI infrastructure software and services and Dell AI-optimized solutions showcased at SC24
- Penguin and Dell Technologies co-sponsor University of New Mexico team in SC24 student cluster competition
Fremont, Calif., – November 18, 2024 – Penguin Solutions™, Inc. (Penguin Solutions, Nasdaq: PENG), today announced its collaboration with Dell Technologies to deliver complete AI solutions that accelerate customers’ time to value through validated, high-performing AI infrastructure. These end-to-end Penguin OriginAI® solutions combine Dell PowerEdge AI-optimized servers, Dell PowerScale storage and Dell PowerSwitch data center switches with Penguin’s validated solution architectures. When enhanced by Penguin’s ClusterWare® intelligent cluster management software and supported by Penguin’s expert AI and high-performance computing infrastructure services — GPU cluster deployment and management are simplified while driving increased performance and availability.
“Rapidly deploying AI infrastructure at scale is a challenge for any organization. Even when successfully deployed, GPU clusters can experience inconsistent performance and availability, eroding the impact of the AI factory,” explained Pete Manca, president of advanced computing at Penguin Solutions. “Dell’s tailored infrastructure platforms, coupled with Penguin’s field-proven AI infrastructure services and software, decrease risk, complexity, and time-to-value for our customers, who are freed to focus on harnessing the value of AI.”
OriginAI solutions are built on Penguin’s extensive experience gained from deploying and managing more than 85,000 GPUs for customers in environments that are characterized by very high levels of node availability and overall cluster throughput. The OriginAI infrastructure solution is being adopted by cloud service providers as well as enterprise customers across financial services, energy, media and entertainment, research and academia, and others in need of reliable infrastructure to support AI workloads at scale.
Penguin designs, builds, deploys, and manages customers’ AI and accelerated computing infrastructures at scale based on best-in-class technology. Additionally, utilizing Penguin’s Scyld ClusterWare® software platform, enables enterprises the ability to monitor and predict cluster health and simplify cluster management.
The OriginAI solution is available with select Dell infrastructure products:
- Dell PowerEdge XE: a family of servers optimized for acceleration and purpose-built for artificial intelligence, generative AI, and high performance computing (HPC). In particular the Dell PowerEdge XE9680, part of the PowerEdge XE family of AI optimized servers, is a high-performance server delivering no-compromise AI acceleration to demanding AI, machine learning (ML) and deep learning (DL) model training, as well as powerful HPC modeling and simulation initiatives with ease.
- Dell PowerScale F710 All-Flash storage: a scalable unstructured data storage platform that delivers superior performance, capacity and efficiency for AI workloads across data center, edge and cloud environments.
- Dell PowerSwitch Z Series switches: a family of flexible, multi-rate high-performance network switches powered by Dell’s highly scalable and flexible Enterprise SONiC distribution featuring generative AI optimized features enabling future-ready data center and AI fabrics.
Penguin’s OriginAI solution will be on display during the Supercomputing 2024 conference in Atlanta, November 18-22 at Penguin booth #1523. For more information, visit the Penguin Solutions SC24 information page.
Penguin and Dell Sponsor SC24 Student Cluster Competition
This year, Penguin and Dell will sponsor the University of New Mexico (UNM) team for the SC24 Student Cluster Competition. This non-stop, 48-hour competition takes place on the exhibit floor where eleven student teams will put their skills to the test against real-world HPC challenges. The UNM team will be supported by Penguin and will use Dell solutions to design and build a cluster that will undergo GPU-intensive benchmarks within a 4500W power limit. For more about the talented UNM team and this year’s HPC Challenge, visit SC24 Student Cluster Competition.
