Be Ready for Tomorrow with an Agile, Enterprise-Grade Data Center Today

Get Power and Flexibility with Best-in-Class Hardware Optimized for Software-Defined Storage

Scalable, Configurable Solutions to Build a Better Data Center

With organizational needs changing so quickly, your data center must be agile enough to address new challenges and innovative enough to incorporate new best practices—all while adhering to strict budgets. However, most proprietary, vendor-locked storage appliances simply can’t do all that.

That’s why the engineers at Penguin Computing™ created the FrostByteTM line of storage solutions. This series of customizable, turn-key software defined storage (SDS) systems uses best-in-class hardware and optimizes for performance and reliability. These highly scalable architectures offer configurable capacity, data-protection, and simple administration for mission-critical data backup, archives, disaster recovery, and replacing older, less efficient servers.

The various FrostByte configurations were purpose built to address the unique data access and storage needs of high-performance computing (HPC), big data analytics, life sciences, financial services/trading, as well as media and entertainment.

To increase performance even further, the FrostByte line is available with expert support and SysAdmin-as-a-Service. To help stretch budgets, all FrostByte configurations are also available as a fully-hosted Storage-as-a-Service.

Case Study: Gemini Man Storage Solution

The Penguin Computing team is proud to have worked on the feature film, Gemini Man, to help director Ang Lee realize his ambitious vision by providing a custom HPC solution. Learn more about how we helped achieve the impossible.

Choose among the following configurations:

Scalable All-Flash Storage for I/O Intensive Workloads

The FrostByteTM with WekaFSTM software-defined storage platform is built on the Relion XE1112 Intel Xeon Scalable processors. With Eight NVMe Drives per node, FrostByte with WekaFS combines the performance of Weka’s SDS platform and the density of Penguin Computing’s underlying hardware architecture.

WekaFS is the world’s fastest and most scalable file system ideal for today’s data-intensive applications in AI and technical computing. It has proven scalable performance, delivering over 10x more performance than blade-based all flash scale-out NAS and 3x more than locally attached NVMe SSDs. WekaFS can help solve any high performance problem, and is ideal for all extreme technical compute and high I/O intensive workloads including machine learning, GPU acceleration, genomics, life sciences, financial analytics, log management and technical compute.

WekaFS was built from the ground up with an optimized network stack that is designed to take advantge of high speed Ethernet or Infiniband networks, so data locality locality is no longer a necessary factor for performance. WekaFS presents a POSIX file system that distributes both data and metadata across the entire storage cluster to ensure massively parallel access. Using NVMe flash technology ensures the highest performance and lowest latency.


Cost-Effective, Petabyte-Scale Object and Block Storage

Red Hat Ceph Storage is an award-winning, production-ready implementation of Ceph that is optimized for large installations (typically a petabyte or greater) and offers both object and block storage, ideal for data lakes, elastic media storage, backup and recovery, and virtualization platforms, like OpenStack®.

Red Hat Ceph eliminates single points of failure and protects data through dynamic data distribution. The system is both self-healing and self-managing, and this reduces administrative overhead costs.

Red Hat Ceph Storage is designed to provide the data protection, reliability, and availability required for demanding, web-scale object storage and flexibility to support a wide variety of workloads and use cases.

Flexible, Scalable File Storage

Red Hat® Gluster Storage is a fast, simple, scale-out file system, ideal for replacing the traditional network-attached storage (NAS) system. It delivers a continuous storage fabric across physical, virtual, and cloud resources so organizations can transform their big, semi-structured, and unstructured data from a burden to an asset.

It is optimized for storage-intensive enterprise workloads such as archive and backup, rich media content delivery, enterprise drop-box, cloud and business applications, virtual and cloud infrastructure storage, as well as emerging workloads such as persistent storage for containerized applications and big data Hadoop workloads.

Speed and Scale for I/O Intensive Workloads

BeeGFS, developed at the Fraunhofer Center for High Performance Computing in Germany and delivered by ThinkParQ GmbH, is a leading parallel file system designed specifically to deal with I/O intensive workloads in performance-critical environments.

This flexible, scalable storage already powers hundreds of scientific and other organizations worldwide. BeeGFS is ideal for demanding, high-performance, high-throughput workloads found in HPC, life sciences, deep learning and cognitive computing, big data analytics, media & entertainment, financial services, oil and gas, and other data intensive analytics.

BeeGFS is natively supported in the Linux kernel and runs on x86_64, OpenPOWER, ARM64, and other architectures. It supports multiple networks with dynamic failover and provides fault-tolerance with built-in replication and file system sanity checks. BeeGFS includes both command-line and graphical tools, simplifying administration and monitoring. And, with BeeGFS On Demand (BeeOND), users can gain the unique ability to create temporary, parallel file systems on a per-job basis.

Gold Partner

Related Solutions