• slide
  • slide


Shattering Expectations

Envision your applications driving previously unimaginable results for analytics, HPC, and other data intensive workloads by using the Pavilion HyperParallel Flash Array™ to deliver unprecedented high performance, ultra-low latency, density, scalability and flexibility, enabling you to achieve things you never before thought possible.

Industry Leading High Performance, Ultra-Low Latency, and Density

The Pavilion HyperParallel Flash Array enables your applications to take advantage of up to 120GB/s of read and 90GB/s of write performance in a single 4RU array. In addition to massive performance, the HyperParallel Flash Array delivers incredible low latency of as little as 25µs. That latency measurement includes not just the array, but storage fabric as well, to deliver the same or better performance than DAS with all the benefits of shared storage.

Dramatically reducing the cost and complexity of your data center, the Pavilion HyperParallel Flash Array packs all that performance into an incredibly dense 4RU array with up to 2.2PB of capacity. Scale up to a full rack of 10 arrays and get 22PB of capacity and up to 1.2TB/s of performance per rack.

For file workloads, every HyperParallel Flash Array can deliver up to 90GB/s of read performance for NFS. Take advantage of fast high performance object workloads with 80GB/s of read performance in a single array. When scaled across an unlimited number of arrays, both NFS and S2 workloads deliver consistent predictable performance of 60GB/s read and 40GB/s of write performance, from each array.

Unlimited, Linear Scalability

The Pavilion HyperParallel Flash Array makes it easy to scale performance or capacity, independently and linearly. Start with as few as 18 drives and grow your storage up to 72 drives for over 2.2 PB in a single array. Begin with only two cards, each of which have two independent controllers, and grow up to twenty controllers in a single array.

Then scale without limits across multiple arrays. Grow from 2.2PB in one array to a rack of ten arrays providing up to 22 PB of capacity. Then scale across multiple racks to support any size data set.

Unified Block, File, and Object

Simplify your management while saving time, money, and rackspace by simultaneously running all your applications on the Pavilion HyperParallel Flash Array providing unprecedented organizational choice and control. Capable of supporting block, file, and object stores simultaneously and with consistent, predictable high performance, the Pavilion HFA lets you power your applications from a single solution.


  • The Pavilion Hyperparallel Flash Array

    The Pavilion HyperParallel™ Flash Array is shattering expectations with industry leading performance, ultra-low latency, density, scalability and flexibility, which enables customers to fundamentally redefine expectations for what they can do with their data, while reducing cost and complexity.

    ●Unlimited scalability across arrays

    ●Native S3, iSCSI, and NFS (NFS v3, NFS v4, pNFS, RDMA) support

    ●Up to 2.2PB per 4RU array, or 22PB per 10 array rack

    ●Up to 1.2TB/s of read performance from a single rack

    ●Concurrent, high performance support for Block, File, and Object workloads

    ●Up to 20 Storage Controllers and Parallel Architecture Boost Performance

    ●40 Ethernet or InfiniBand Ports for Flexible Connectivity

    ●Up 120/90 GBs R/W Throughput, 25 µs Latency, and 20M 4K Random Read IOPS

    ●Active/Active Data Access

    ●No Single Point of Failure

    ●End-to-End NVMe with simultaneous NVMe/Ethernet, NVMe/InfiniBand, and NVMe/TCP protocols

    ●Up to 72 U.2 2.5” NVMe NAND and SCM SSDs

    ●NVMe-oF/RDMA (Ethernet and InfiniBand)


    ●100% Standards Based

  • Pavilion HyperOS 3.0

    Pavilion HyperOS™ 3.0 powers the Pavilion HyperParallel Flash Array™ and together, the Pavilion HyperParallel Data Platform™ delivers organizations with the highest performance, ultra-low latency, density, scalability, and flexibility in the industry. With support for block, file, and object protocols, Pavilion HyperOS shatters performance expectations to enable your applications to do more for your organization than ever before.

    Built to unlock the power and performance of NVMe and NVMe-oF, Pavilion HyperOS enables the Pavilion HyperParallel Flash Array to deliver consistent, predictable, high performance, ultra-low latency, and linear scalability. Designed for the most demanding environments, Pavilion HyperOS combines enterprise class management, security, and data protection features with a highly intuitive GUI to ensure maximum availability and ease of use. An API driven approach ensures that the Pavilion HyperParallel Flash Array easily integrates with your ecosystem for simple management.




May 01, 2015- Apr 30, 2025


Archived Events


Latest News

Pavilion Data Systems has been recognized with a Flash Memory Summit 2020 Best of Show Award for the Most Innovative Flash Memory Customer Implementation during the Flash Memory Summit 2020 Best of ...
Pavilion Data Systems, Inc. and Carahsoft Technology Corp., The Trusted Government IT Solutions Provider®, today announced a partnership.



Learn how Organizations achieve unprecedented Spectrum Scale Performance. This analyst paper shares how Pavilion solutions optimize GPFS / Spectrum Scale Deployments.


Pavilion Datasheet
With support for block, file, and object workloads at up to 120GB/s and 20M IOPs in a 4RU footprint,the Pavilion Hyperparallel Flash Array is the industry’s most capable and highest performant solution for delivering real time performance at scale within reach of all.

Pavilion Datasheet
PavilionOS combines enterprise class management, security, and data protection features with a highly intuitive GUI to ensure maximum availability and ease of use.

Pavilion Datasheet
Pavilion’s Customer support offers a range of services that enables our customer to achieve their business goals and fully leverage their investment.


Pavilion Resources
Learn about Pavilion HyperOS 3.0, Powering the Most Performant, Dense, Scalable, and Flexible Storage Platform in the Universe

Learn about Pavilion’s a speedy parallel file system that brings it into direct competition with major HPC storage players for the first time.

A souped-up parallel file system by Pavilion Data Systems brings dense, distributed multiprotocol flash storage capabilities to supercomputing workloads.


Learn how Pavilion is revolutionizing data centers by shattering expectations across the most critical measurements of performance, density, scalability and flexibility.


Attendees of this webinar learned how to accelerate critical applications performance, and how federal agencies are deploying NVMe over Fabrics hyper parallel storage solutions.


Pavilion Whitepaper
Learn why modern applications suffer from poor storage performance density and why the common storage scale-out server centric architecture of software defined DAS storage (SDS), all-flash-arrays (AFA), and even shared accelerated storage fail to provide the necessary storage performance density...